August 1, 2025 • 50 mins
Today we speak with China military scholar Elsa B. Kania about China’s military modernisation. How good is the People’s Liberation Army? Where has it progressed? Where is it still deficient? And the big ones: can it match the US and how ready is it to take Taiwan by force if Xi Jinping gives the order?
Much of Elsa’s recent work has focussed on the role of technology in the PLA’s capabilities, doctrine and command structure. She talks about the role of artificial intelligence, the concepts of informatisation and intelligentisation, and the Chinese view of the ethics of automating lethal force. She also talks about China’s military rehearsals around Taiwan, its concept of “peace disease”, and China’s overall strategy with its growing military assertiveness.
Elsa is a PhD candidate in Harvard University's Department of Government, where she’s just recently defended her dissertation, "China's Command Revolution." Her research focuses on China's military strategy, defense innovation, and emerging capabilities. She is an Adjunct Senior Fellow with the Center for a New American Security's Technology and National Security Program, and she was also a Fulbright Specialist and Non-Resident Fellow with the International Cyber Policy Centre at ASPI.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:01):
Stop the world. The PLA is definitely actively
experimenting. There's some claims that some
tools have been deployed. There's definitely been
demonstrations of swarming, claims of weapon systems with
different degrees of autonomy, claims that decision support
(00:21):
tools enabled by AI are being deployed, including at the
theatre command level. Welcome to STOP the World, the
SP Podcast. I'm Olivia Nelson.
And I'm David Rowe. Dave, I love today's guest, Elsa
Kenya. She is brilliant.
She's smarter than my phone. I'll also bet that she's smarter
than GPT 5, which is rumoured tobe coming out this month.
(00:44):
She's damn smart anyway, and we're grateful that she's
devoted her efforts to understanding the Chinese
military. We all know it's made huge
progress in recent decades, and in terms of quantity, it's
outmatched in the US in some areas.
But just how good is it, especially not having fought a
war in decades? Yeah, so that's really the
debate that military experts around the world are having, and
Elsa knows as much about it as anyone in the public realm,
(01:07):
especially on the ways the People's Liberation Army has
incorporated technology. Elsa is a PhD candidate in
Harvard University's Department of Government, where she's just
recently defended her dissertation, China's Command
Revolution. She is an adjunct senior fellow
with the Centre for a New American Securities Technology
and National Security Programme,and she was previously a
(01:30):
Fulbright specialist and non resident fellow here at ASPY.
When I spoke with her this week,she was recovering from running
a half marathon live. And I was wondering what exactly
I'm doing with my life. Me too, Dave.
So Elsa talks about how the PLA matches up against the US
military, the recent history of its improvements, the concepts
(01:51):
of informatization and intelligentization, the
centrality of AI to its modernisation, and the Chinese
view of the ethics of automatinglethal force.
So it's quite a mouthful, but you got it out live well.
Yeah. Thanks, Dave.
She also talks about the significance and purpose of
China's military rehearsals around Taiwan, its concept of
peace disease, which translates as its own lack of real world
(02:12):
fighting experience in China's overall strategy with its
growing military assertiveness in various arenas.
There's a lot packed into this conversation, but it rolls along
and really, Elsa is just a greatsource of insight into this
pretty murky area. So let's take a listen.
Welcome to STOP THE World, everyone.
I'm here with Elsa Kenya. Elsa, thanks for joining me.
(02:33):
Thanks so much, happy to be here.
So let's start by looking at a high level at the progress that
the PLA has made in recent years.
The US has deemed it a pacing threat for a while now.
I mean, we don't have to go through the whole history, but
as a way of giving an overview to the conversation perhaps, how
would you quickly characterise the state of the PLA?
(02:56):
Where has it made the most advances since, say, 2000?
Where is it still deficient? And just in headline terms, you
know, where is it strongest? Sure.
So the PLA has achieved dramaticprogress but continues to
confront considerable challengesand I'd I'd point to significant
(03:21):
expansion of Chinese cyberspace capabilities as well as space
and counter space. There's the PLA rocket force of
course and dramatic developmentsand hypersonics as well.
But I think some of the deficiencies may persist in the
lack of operational experience, questionable capabilities of
(03:42):
Chinese military commanders, thepotentially some improvement on
that front given attempts to upgrade their training and their
education and cultivation of talent.
So I think assessing Chinese military power is inherently
complex and uncertain. I tend to think that the truth
lies somewhere in the middle, between those who would dismiss
(04:05):
the PLA as a military museum andthose who would argue it's
already leagues ahead of the US military.
It's hard to right size our understanding of Chinese
military power, but I think it'simportant to both recognise the
progression but also understand that the PLA still sees itself
as striving to catch up with theUS military, which it regards as
(04:27):
a powerful adversary and the benchmark for its own
modernization. And there is still a level of
anxiety about the US military moving ahead into new
capabilities, pursuing new technologies.
There is a fear of falling behind or of technology
surprise. But at the same time, there does
(04:48):
appear to be a greater degree ofconfidence today that the PLA is
closer to closing that gap, especially when it comes to new
domains and emerging capabilities, whether that is
cyberspace or artificial intelligence or exciting
developments in quantum technology as well.
Depending on the time horizons, different capabilities may have
(05:10):
different degrees of relevance in in various contingencies.
But the PL as ambition is clear to not just pursue a strategy of
catching up, of developing asymmetric capabilities to
counterbalance the US military, but now looking as well to
achieve the leapfrog development, so to speak, and of
(05:32):
genuine disruptive innovation and trying to achieve an
advantage or the strategic commanding heights, as Chinese
military commentators often phrase it, in some of these new
domains and technologies where the US does not have a clear or
decisive advantage. So I I think we should be
sceptical sometimes of the claims in Chinese military media
and propaganda, but also take these competitive challenges
(05:55):
quite seriously. I mean, this is a layperson's
question, I suppose, but is it possible to put them side by
side and say the PLA is, I don'tknow, 70% of the way there, 80%
of the way there, something likethat?
I am hesitant to put a number onit in the sense that it's apples
and oranges in some sense for lack of a better metaphor.
That talking about two militaries with very different
(06:17):
histories, different ambitions and optimise for different
scenarios. Where U.S. military has been
focused on power projection and has recent experience in the
Middle East primarily which has varying degrees of relevance
contingencies in the Indo Pacific depending on one's
vantage point. Whereas the PLA does not have
(06:38):
that contemporary operational experience and is very focused
instead of on being a global military power, on regional
contingencies, specifically Taiwan, and sort of first on
counter intervention campaigns on preventing U.S. military
projection of power. So it's almost thesis and
antithesis in some sense where we're looking at different
(07:00):
paradigms of military power. Of course, we've also seen the
PLA try to emulate some of the hallmarks of U.S. military
power, as in the development of aircraft carriers.
But the PLA is focused much moreso on asymmetric capabilities
and on counter intervention and developing capabilities to
target the weaknesses it recognises in US paradigms of
(07:22):
military power. Yeah.
OK, Now that's fair enough. Some of your work in the work of
others has looked at how they'veput information at the heart of
their modernization. Can you just talk a little bit
about that? I mean, in what sense has their
focus on information really kindof, I suppose, shaped their
modernization? So I'd say that the PLA, since
at least the 1990s has really concentrated on information as a
(07:47):
central feature of military power, focusing on information
power as an enabler of combat capability and seeking to
achieve information dominance inthe early stages of any conflict
scenario. And I'd say that is critical to
how the PLA regards the former character of conflict.
(08:09):
And this focus on information has driven changes in Chinese
military strategy and their doctrine and sort of core
concepts at the heart of how thePLA is preparing to fight.
And we can look back at the lessons the PLA learned from the
Gulf War and from US conflicts in Iraq and Afghanistan and
(08:30):
looking at sort of network centric warfare and trying to
develop their own approaches focused on what they call
informationization. Xin XI Hua, which really is this
transformation of warfare and ofmilitary power through
information. And that has driven the
development of capabilities for information operations, namely
(08:52):
cyber, electronic and psychological warfare focused on
targeting their adversaries, decision making and and
targeting vulnerabilities in battle networks as well.
And then also developing greaterresiliency in their own C4 ISR
capabilities, their own command information systems and trying
to ensure a degree of resilienceand redundancy having seen how
(09:16):
the US and its conflicts has often targeted those sorts of
networks and systems. So this is also closely linked
to concepts of sort of system ofsystems confrontation as a core
feature of modern and future warfare.
And increasingly, beyond this emphasis on informatization,
(09:36):
there's also this focus on the impact of artificial
intelligence as a new catalyst for changes in the form of
warfare, which is leading to these these emerging and
evolving debates on intelligentization or the
transformation of warfare based on advances in AI and related
emerging technologies that are seen as an element of a
(10:00):
revolution in military affairs. And US debates on the revolution
in military affairs sort of dwindled somewhat after the
early 2000s, whereas the PLA hasconsistently seen this framing
of military revolution or transformation as really pivotal
to how they are approaching, initially catching up in
advances in information technology and today trying to
(10:22):
achieve a leading position and new frontiers of artificial
intelligence. Yeah, yeah.
And I definitely want to come back to AI and B
intelligentization. So I mean, from what you're
saying, I mean they're looking at it both from a point of view
of information warfare in terms of, of operations looking for
vulnerabilities in adversaries, I suppose information networks
in order to potentially disrupt those.
(10:44):
But also information, I suppose harmonisation within their own
forces to, you know, improve their ability to fight as a sort
of coordinated and joint force. If I, if I understand it
correctly, I'm interested in, you know, the third of those
categories, what it's meant for their command and control.
I mean, that was, I think their C2, as it's called, was
(11:05):
traditionally seen as a bit ropey.
They had difficulty with, you know, I suppose the the lack of
initiative at shown at more junior levels and the the
strategic goals set by the the higher levels making their way
down. And so basically their command
and control was well below wherethey wanted it to be.
What has this information utilisation, if that's the word?
(11:27):
What has that meant for their command and control in
particular, and how good is it now?
So I guess I'll start with some of the self criticisms you
referenced that dating back to at least 2015, Chinese military
leaders and commentators, including up to the level of Xi
Jinping and sort of senior members of the Central Military
(11:48):
Commission were quite candid in pointing out what they saw as
shortfalls in the commanders capabilities.
The so called 5 incapables, including that commanders at the
time were quote unquote. Some commanders had difficulty
understanding the intentions of leadership, of judging the
battlefield situation, of decision making and uncertain
(12:10):
situations and so on. And that this this actually was
the driver for new approaches totraining focused on trying to
overcome these self assessed shortcomings and to ensure that
commanders could be more capableof exercising their subjective
(12:30):
initiative and interpreting senior leadership guidance,
including understanding the party's intentions and being
able to respond to more complex and dynamic battlefield
situations. So I think when we hear these
self criticisms and often sort of a tendency to echo these in
in debates among external observers and how to right size
(12:51):
Chinese military power. I think it's important to note
that the PLAC has been very forthright in discussing these
problems and has been very targeted and focused in its
efforts to overcome some of these challenges in these human
elements. And certain more recent
(13:12):
commentaries still allude to thefive incapables and other self
criticisms as not fully resolved.
But I'd say in the past decade there has been a certain amount
of progress or some indicators that the more realistic
training, the effort, efforts toreform elements of their
(13:32):
military education and cultivation of joint commanders,
as well as grassroots commanderstrying to encourage them to be
more creative and adaptive are starting to gain some traction.
So I do think it's important to recognise that the PLA is a
learning organisation and they are adapting.
They are trying to overcome the weaknesses that they recognise
(13:53):
that relate to the peace diseasethat they see as a problem given
their lack of contemporary operational experience.
And there are reasons to be sceptical that they can fully
overcome these problems. But I think it's important not
to discount the possibility thatthey may achieve greater
progress than we may appreciate.Though of course, for those of
(14:14):
us who watch the PLA, based on PLA media and military writings,
there's always the caveat that so much of the information we
rely upon is generated from within the system.
So we see a level of candour, but there's also can be a
certain amount of distortion andobviously controls and
censorship and propaganda associated with some of this
(14:37):
representation as well. So I think it's important to be
to be sceptical, but also recognise that there the PLA has
invested and aligned its training to overcome some of
these previously identified deficiencies and commanders
capabilities. So that's the human element of
it. From a technical perspective,
(14:58):
there have been significant investments in the resilience
and redundancy of their whole architecture of command control
communications, as well as the information surveillance and
reconnaissance capabilities thatconstitute elements of the
information support that would enable precision strikes.
(15:19):
And the stand up of the Information Support Force in
April of 2024 based on an elevation of elements of the
former strategic support force, I think reflects the importance
and prioritisation associated with continuing to advance their
command information systems and the complex network information
systems that are recognised as critical capabilities for
(15:42):
current and future conflict. So say that there have been
major advances in China's deployment of space based
systems for surveillance to support strikes and targeting.
There's been very redundant, robust ground based architecture
(16:02):
of information networks to enable insured communications
and continued attempts to innovate and integrate new
technologies into this overall architecture of command and
network information systems, focusing on improving this
overall capacity for informationsupport and including greater
(16:23):
decision support. So there may be a tendency to
turn to technology to compensatefor some of these human
shortfalls. But I think inherently there's
also a pragmatic recognition that the complexity of
contemporary operations is such that no human commander, no
matter how capable, can make decisions effectively unless
(16:45):
they have the right information and the right analytic
capability and the right intelligence.
So I think information support and greater intelligentization,
so to speak, in the processing of ISR and use of some of these
new tools and techniques to support command decision making
is also critical. So this focus on improving the
(17:05):
human element and upgrading the technology are proceeding in
parallel. Yeah.
OK. Now, I was going to ask about, I
mean, that parallel dimension isa really interesting one.
I mean, the technology has improved, but at the same time,
I suppose their philosophy, their doctrine, what you will,
has shifted at the same time towards this approach of let's
make sure we can see what we need to see, share the
(17:28):
information we need to share in order to be able to make good
decisions, give and receive the orders that we need to give and
receive. And that's turned them from what
you're saying into into a very aware, very commutative and and
potentially very decisive force overall.
Yeah. Absolutely.
And I do think that there is this focus on integrating and
(17:50):
reconciling these human and machine or technological
dimensions of warfare. I've been asked a number of
times over the years, do I thinkthe PLA will take humans fully
out of the loop and have machineintelligence commanding all
operations. And I tend to be sceptical of
that. But I think there, I always say
I'm highly confident in the nearterm at least one human will be
(18:11):
in the loop for all strategic decisions.
That would be Xi Jinping personally, his role as
commander in chief. But when it does come to more
tactical and technical operations, there appears to be
a fairly flexible and pragmatic approach to leveraging emerging
technologies for applications they're uniquely suited for.
So for instance, greater automation of elements of
(18:34):
cyberspace operations, for instance, including uses of
generative artificial intelligence to accelerate and
enable scaling up of some of these capabilities and
exploitations I think is something we should expect to
see and some evidence there's experimentation along those
lines. Applying AI tools, including at
the edge to accelerate processing of intelligence,
(18:58):
including from some of their space systems is something we
can expect to see even some start UPS launching what they're
describing as AI enabled satellites.
So think a lot of very technologically deterministic or
solutionistic approaches to leveraging technology where it
has certain strengths. But there also appear to be
(19:19):
fairly nuanced debates within the PLA about the strengths and
weaknesses of machine intelligence relative to human
intelligence and the recognitionthat there can be certain biases
or issues of safety or surety ofthese systems, which
necessitates A continued human involvement and command decision
(19:39):
making at the operational strategic levels.
Even if certain tactical applications, elements of air
and missile defence, for instance, or aspects of
swarming, suicide, drones and and the like could be more fully
automated. So I think there's not a single
answer or a single approach in terms of how the PLA may use
(19:59):
these tools or may approach integrating technologies.
The lot of this will depend on different domains, applications
and the evolving maturity of these capabilities.
As you've sort of anticipated where I wanted to go next in a
wonderful way, I mean that that is a really just a fascinating
insight into the into the way they're adopting AI and
(20:20):
automation. And at this point, I just want
to give you a quick shout out for the 2019 Aspy report that
you wrote on Chinese Russian cooperation in areas like AI.
So fully six years ago, not justwere you anticipating the China
Russia cooperation, which of course has become one of the
biggest issues the strategists are considering in 2025, but
(20:40):
also the importance of things like AI and other advanced
technologies. So a quick shout out there, but
just, I mean, that's thrown up an ethical question for me.
I mean, it's one that's as old as the hills, but the basic idea
of using lethal force without a human in the has generally been
seen as a red line for militaries who are answerable to
democratic systems. Do you think that is not an
(21:02):
ethical red line for the PLA? So my impression would be that
the PLA is very pragmatic in their approach.
There are debates in Chinese military writings about ethics,
about law of armed conflict, about the benefits or drawbacks
(21:23):
of having a human in, on or out of the loop.
But with the caveat that these days discussing directly with
Chinese military counterparts ismore challenging, my impression
of these debates is that the PLA's approach is relatively
(21:45):
pragmatic. And I suppose I keep coming back
to that characterization becausethere does appear to be a degree
of concern with law of armed conflict, legitimacy and use of
force, a desire to have precise and accurate targeting as
opposed to great risk of collateral damage.
(22:07):
And the PLA is a professional military.
They're not looking to cause mass casualty events without of
control autonomous weapons. They want to ensure that their
weapon systems that may have various degrees of automation or
autonomy built in will hit the targets they intend for them to
hit. So I do think that there is a
(22:30):
active debate or there appears to be ongoing consideration of
the of the relative advantages and disadvantages of having
human decision makers in on or out of the loop at different
echelons for different instancesof employment of weapon systems.
But I think the notion that a human has to be pulling the
(22:53):
trigger, so to speak, metaphorically appears to be
less of a moral red line. And I think that there is.
Without weighing into these debates in US or Australian
context, I think there is something to be said for a
recognition that simply having ahuman quote, UN quote in the
loop is not enough for the decision to be an ethical 1.
(23:16):
I expect these will continue to be active debates in across
multiple militaries as the technology evolves, as the
calculus of risk evolves. And much may depend as well on
the operational environment. And the types of decisions or
capabilities that could be employed, for instance, in urban
(23:37):
warfare would differ greatly from the considerations if we're
thinking about naval warfare or maritime operations.
So I think the level of complexity and congestion of the
environment may be a factor. And I suspect the debates will
continue evolving in the PLA as here.
But the overwhelming considerations driving the PLA's
approach appear to be pragmatism, and the question of
(24:02):
which approach is technically and operationally advantageous,
as opposed to more abstract notions of red lines morally.
Yeah, yeah. No, you're right.
It's far from a simple discussion.
You wrote recently about the sort of jolt that China and and
including Xi Jinping personally got from the famous AI battle
(24:24):
between Alpha Go and the then Goworld Champion Lisa Doll and the
trajectory. I mean, famously Alpha for Go
defeated Lisa Dole, much to everybody's surprise.
And it was a huge moment really well across all of Asia, where
go was this ancient and much loved and much played game,
probably missed in much of the West other than in circles that
(24:45):
were paying a lot of attention to to AI.
But you described recently that that sort of, you know, set
China on a trajectory. It was a sort of AI hate the
term Sputnik moment, but it was a bit of a Sputnik moment for
China. Just talk us through that a
little bit in the trajectory that it set China on in terms
terms of, I suppose, building AIcapabilities for including for
(25:06):
military purposes? Sure.
So looking back at Alphago and looking at the Pele's reaction
at that time, there did appear to be a catalytic impact to some
extent in terms of the recognition that this was a
demonstration of how quickly advances in AI were progressing
(25:27):
and the fact that an AI system could defeat the world's top
human players in a game. That did centre on decision
making, stratagems and complex calculations.
And think the salience of the game of Gove to some extent may
have been a factor there. But also the notion that during
(25:48):
some of these matches with Lisa Dole and other players, Alpha Go
started to introduce moves or plays that humans hadn't
conceived of. And the totality of the history
of the game of Go. So I think that.
The famous move 37, I think, wasn't it where, where everyone
was so aghast and and everyone thought that that the computer
had made it a terrible, disastrous mistake.
(26:09):
And it turned out to be a a brilliant strategy that, like
you say, no one had ever used before in the in the human
playing of the game. Indeed, and as someone who was
watching PLA debates and writings at that time and
initially having looked back, and this is history now, but the
PL as responses to the third offset strategy that was
launched around 2014 and through2015 when I was initially
(26:33):
started tracking these issues. The PLA both saw the strategic
and operational potential of these developments in the
context of their concerns about U.S. defence innovation
initiatives and believed there could be unique utility to
integrating AI into future command decision making.
(26:56):
And this prompted some of the early debates that had harkened
back to the PL as earlier interest in thinking about how
technology could enable an upgrade command decision making
on command as a complex system with human and machine elements
(27:16):
integrated. And this this strive to have a
more scientific approach to warfare and to command decision
making. And Alphago, in that context,
did spark a lot of debate and some apparent initial efforts
and experimentation with integrating AI and developing
tools for decision support. There's also a lot of efforts in
(27:38):
war gaming coming out of institutions like the Chinese
Academy of Sciences, the Palais National Defence University,
National University of Defence Technology.
A lot of focus on trying to develop AI agents that could
play war games and both AI versus human and AI versus AI
(27:59):
machine versus machine confrontation as means of
testing and validating how effectively AI tools or agents
could be used in military decision making, at least in a
games play and context. Of course there's a great
difference between a even the most sophisticated of
computerised war games relative to the complexity of an actual
(28:20):
real world operational environment.
But this was at least seen as a stepping stone or proof of
concept to validate how some of these AI agents or systems could
start to be developed and how they could be paired with human
players and that Centaur type approach to decision making as
well. So I do think that this was one
(28:42):
of several moments in the PL as recent history that did draw
their attention to the potentialof AI.
And I do find it ironic that this this moment was not seen
very much as of military relevance in AUS or generally
western context, but did spark acertain amount of debate in the
PLA, as did more recently their reaction to Chechnya PT and
(29:06):
concerned that at a moment whereChinese leaders and observers
have started to be more confident, they were closer to
drawing even with USAI developments.
ChatGPT, I think was another another moment of shock again,
I'll, I also hesitate to say a Sputnik moment, but certainly a
moment that had some catalytic impacts in driving efforts to
(29:27):
think about how large language models and generative AI
generally could contribute to future military capabilities,
whether in in intelligence and command decision making or in
other aspects of planning and strategizing.
So there is a a degree of reactivity and how the PLA has
(29:47):
responded to global developmentsin AI and how closely they've
monitored advances, especially in in AUS context, concern
always about the US as a powerful potential adversary and
fear of this fear of technology surprise.
That's a consistent theme. But there's also efforts to
beyond scanning global military and technological developments
(30:10):
to try to anticipate where some of these trends are going.
And not just to react and be a fast follower, but to be more of
a leader in original innovation and to design future warfare or
or design the parameters for thefuture battlefield.
So the PLA is I think a ways to go and truly achieving that.
But there's certainly an ambition and very close study of
(30:33):
technology and of some of the theoretical implications as
well. Yeah, well, we'll try not to
blame DeepMind and Open AI too much for catalysing China's
military modernization efforts. Is there a way to, in an apples
to apples sense, say who is doing a better job between the
PLA and the US military at actually integrating AI,
(30:58):
including in a human machine teaming sense?
That is a fantastic question andI don't have a great answer.
I think it is if. Anyone has it you do.
Well, it is an aspect of ongoingresearch for me.
I look more at the PLA side of the equation, so I'm less
confident making a comparison since I'm less fully familiar
(31:22):
with the US context here. But I think, I guess I'd say by
way of caveat, I think for both militaries, we're still at a
stage where there is a lot of experimentation.
There are a lot of efforts to have proofs of concept to try to
test and validate what systems or use cases and applications
(31:42):
can be feasible and advantageous.
There's a lot of very diverse efforts across every application
imaginable. Just about.
And at the risk of sounding cynical or flippant, I think
that sort of just at AI is, I think, a tendency sometimes of
military leaders in multiple systems to sort of have this
level of enthusiasm or exuberantexpectations of these
(32:05):
technologies that sometimes start to confront the realities
that when you are trying to implement complex advanced
technologies within military bureaucracies in any cultural
context, there will be obstacles.
And there will be difficulties with how you manage data, how
you upskill your personnel at different echelons to use these
tools effectively. How you start to develop new
(32:29):
procedures and parameters for integrating these technologies
into operations, including in training and exercises and
recognising some of the problemsthat arise.
So I guess I'd say ask me again in a couple years.
I think the PLA is definitely actively experimenting.
There's some claims that some tools have been deployed.
There's definitely been demonstrations of swarming
(32:52):
claims of weapon systems with different degrees of autonomy,
claims that decision support tools enabled by AI are being
deployed, including at the theatre command level.
Lot of applications. I think some of the earliest
applications we'll see will be in virtual domains where it's
harder to measure or observe from the outside, whether that
(33:13):
is in accelerated processing of data to enable enhanced
intelligence capabilities or integrating generative AI for
cognitive domain or cyberspace operations.
So I think some of these earliest manifestations may be
in virtual domains or in some ofthese information related
(33:35):
functions and capabilities. There's been some some
indications of AI adoption in different aspects of cyberspace
operations and that could including for malware
development, social engineering and the like generation of
content for more targeted psychological operations.
So I think think some of the near term applications will
(33:59):
observe in competition or confrontation will be along
those lines how effectively certain capabilities are being
integrated into operational decision making or harder to
observe from the outside. So it's easier to point to the
indications of effort and progress, harder to say how
effective or successful these initiatives are.
(34:21):
And I guess I'd add as well thatthere there is I think a greater
extent of uncertainty about the military balance because these
qualitative advancements could have significant applications
and implications for the future of warfare.
But there is an inherent difficulty in measurement.
It's hard to count or hard to measure how much having a
(34:43):
certain AI modular system deployed enhances targeting if
it's say an automatic target recognition and Chinese
missiles. It's hard to hard to measure or
count or evaluate in the abstract.
And I think this also highlightswhy we need greater integration
of subject matter and technical expertise.
(35:03):
I'll be the first to admit I'm as political scientists by
training, I am always trying to continue learning and upscaling
myself. I'm actually here at Berkeley
for an AI red team in boot camp this week, which has been a
fantastic experience so far. So definitely a proponent of
lifelong learning for all of us to keep pace with this.
But I do think that for those ofus who've been observers of
(35:23):
military affairs, I think the how we calculate or how military
strategists and decision makers calculate the future military
balance may depend more and moreon complex technologies for
which the metrics are difficult and uncertain.
Which I think does also increasethe risk of some kind of
miscalculation or misperceptionsthat could arise with that new
(35:47):
fog of technology, so to speak. I'm envious of where you are now
and I'm just picking up on your point.
I think that, you know, the challenges that you were
referring to there of, you know,just add AI and assuming that
integration and adoption is straightforward is a challenge
that in fact large organisations, military, private
sector, civilian of all kinds finding.
(36:07):
And it's, it's fascinating that one of the, the bottlenecks, as
it turns out, is, is simply organisational readjustment in
order to, to integrate AI. It's interesting, but unless I
go down a rabbit hole of my own digging there, let's just there
are a couple of other things that I do desperately want to
cover off on while we're here. And what 1 is just you've
touched on the PL as lack of, you know, real World War
(36:30):
fighting experience and you've sort of indicated that that is
being overcome in some ways. How relevant or significant does
that remain? And to what extent do you think
that the rehearsals around Taiwan as Admiral Paparo Indo
Picom commander has described them?
To what extent are those actually about, you know,
(36:51):
practical experience and and learning and actually improving
in that real world scenario? The PLA is quite candid in
admitting to the challenges it'slack of contemporary operational
experience can present. The so called peace disease
which is often referenced in commentaries which are often
accompanied by expectations for more combat realistic or actual
(37:16):
combat training. And the extent to which the PLA,
especially the Eastern Theatre Command and its continual
presence and operations around Taiwan can get experience in the
likely operational environment for its priority scenario and
can validate command relations, complex multi domain
(37:39):
synchronisation in some of thesepressure operations.
And that is perhaps at least a partial mechanism of overcoming
that. So this focus on greater
sophistication in training on the integration between
peacetime employment of militaryforces with likely wartime
(38:03):
contingencies that all is aimed at overcoming some of these
factors, as are the emphasis on war gaming simulations and this
fairly scientific approach to trying to anticipate some of the
challenges that could arise in afuture conflict scenario.
(38:25):
Think military observers have been quite wrong in, for
instance, assessments of Russiancapabilities.
So somebody who studies the PLAItend to be quite humble and
cognizant of the limitations in in our community's perspective
and what we see and don't see. But I do think the fact that the
(38:46):
PLA, especially the Eastern Theatre Command, which has been
responsible for orchestrating these complex pressure
operations, has been seemingly successful, at least as
portrayed in their own media andundertaking these named
exercises and major campaigns around Taiwan.
(39:07):
That is seemingly advantageous in their familiarisation, in
their capacity to rehearse as Admiral Papparo's characterised
it, and as contributes to overall readiness.
So I definitely don't think we should dismiss peace disease,
and certainly the PLA does not dismiss this as an impediment,
(39:27):
but so too do our our own militaries not have experiences
that quite align with the complexity of what a future
conflict in this theatre could entail.
I hadn't heard the term peace disease before.
It's a bloody depressing phrase to hear.
Indeed. You you make a good point about
(39:48):
the Russian military there as well, and how it reflects on the
humility we should show in making bold predictions about
any fighting force. OK, are you prepared to say how
red do you think the PLA is to mount a successful blockade or
even invasion of Taiwan at the moment?
So I would say the current posture as apparent and well
(40:13):
documented of the PLA in the vicinity of Taiwan gives them
the optionality to transition into a blockade or quarantine
should they choose to do so. And the strategic calculus or
rationale is something of a question.
But in terms of the the extent to which the PLA has expanded
(40:37):
and attempted to normalise theirposture around Taiwan is
significant and has expanded dramatically, especially since
Speaker Pelosi's visit, which the PLA chose to respond to with
massive pressure operations thatthey justified as a proportional
(40:59):
reaction, but did create a sea change and a step towards
expanding their posture in ways that give them that optionality
and capability to transition into a blockade.
And certainly that has been someof the signalling not not so
(41:19):
subtle in some cases associated with elements of their joint
Sword or Straight Thunder exercise series.
So a blockade or quarantine and there there would almost
certainly be some kind of legal warfare or political warfare
justification accompanying it asan attempt to rationalise it
(41:41):
that would not be all that operationally challenging based
on how they are already postured.
And I think this incremental expansion in attempts to change
facts on the water has been significant.
But that is a far cry from readiness for an actual
invasion. And I do tend to be of the
(42:03):
school of thought that although the threat of an invasion is not
trivial and shouldn't be dismissed, there are many tools
and techniques through which thePLA could escalate their
posture, presence and pressure around Taiwan in an attempt to
(42:27):
deter or intimidate or attempt to force acquiescence prior to
attempting an invasion. Because an invasion would be
incredibly challenging. So the PLA has been evidently
directed by Xi Jinping to be prepared by 2027.
Which again, I think often this point gets conflated in some of
(42:52):
the media coverage and public debates.
But becoming capable by 2027 does not imply an intention or a
deadline or timeline centre in 2027.
But the the benchmark has seemingly been established as
their centenary objective of becoming a powerful military on
(43:13):
a course to becoming a world class force and assuring China's
national rejuvenation by mid century and so on and so forth.
That they are seeking to have the capability to invade,
including as a means of attemptsto coerce Taiwan.
But I think I would suspect thatthe PLA today is quite a ways
(43:36):
from actually becoming confidentit has the ability to do so or a
decision to invade would be a major gamble on the part of
Chinese leadership and there would be non trivial risks in
that choice. So I do think that there are
reasons for concern. We could see orthogonal
(44:00):
approaches or horizontal escalation or different
techniques trying to target Taiwan and also trying to target
US calculus, since I think ultimately Chinese leaders and
observers appear to regard US decisions and U.S. military
(44:21):
power as the primary, if not predominant obstacle to their
capacity to achieve unification as they term it.
So I think we can expect to see more creative techniques and
attempts to impact US calculus or Taiwan's decision making
(44:43):
prior to any actual attempted invasion, but that does not mean
that the threat is not quite serious.
Yeah, yeah. Fascinating.
All right. And now, Elsa, to wrap up, I'm
going to be a bit grossly unfairhere, but I think you're so good
on this stuff that you can do it.
But I'm going to ask you to like, in roughly a minute or so,
(45:03):
taking account of China's military modernisation, its
activity around Taiwan, it's activity in the South China Sea,
the East China Sea, you know, construction of bases further
afield globally. How would you quickly describe
(45:23):
Beijing's overall military strategy in the service of its
broader national interest? So I'd say that when it comes to
activities within the first island chain, East and South
China Seas, Taiwan Strait, Chinese leaders are trying to
(45:46):
transform China into a major andgreat maritime power that has a
dominant position within its near seas, within the first
island chain and can exercise effective control of those
waters. I do.
I believe and I'm concerned thatChinese leaders, including Xi
(46:09):
Jinping especially, regard unification with or de facto
coercive annexation of, if necessary, of Taiwan by mid
century as among the benchmarks for national rejuvenation or the
XI Jinping's speeches and statements have articulated that
(46:32):
objective. And I do think that Chinese
military strategy and capabilities development are
aimed at counter intervention, and namely countering what
Chinese leaders regard as the potential for US intervention
(46:54):
that is seen as an obstacle to achieving these objectives, as
an animating driver of this asymmetric counterbalancing
strategy, as some Chinese military textbooks describe it.
So I do think that there are efforts to achieve incremental
advancement towards these aims. Xi Jinping and some of his
(47:18):
statements and the growing number of commentaries highlight
peacetime employment of militaryforces as an important element
of this approach of as the PLA becomes a more capable military,
the ability to brandish that power and to achieve deterrent
or coercive effects on China's neighbours in ways that
(47:39):
strengthen their position. And some of these territorial
disputes in the region is is emerging as a major feature of
their approach as well. And I do think that ultimately
this is driving to this ambitionby mid century to be a world
class or global first class military, which appears to imply
not only relative dominance in the region, but a transformation
(48:05):
of the military balance. The Chinese strategists and
observers appear to be becoming increasingly confident they are
on track to achieve and that notonly this transformation of the
cross Strait balance, but also adramatic alteration of the US
China military balance that is often accompanied by or often
(48:28):
accompanies some of these claimsthat the overall balance of
power, of comprehensive nationalpower is shifting in China's
favour. I believe there is a degree of
overconfidence in some cases that is juxtaposed with intense
insecurities on on other fronts.But the ambition is clear.
(48:50):
And although Xi Jinping gets a lot of credit for what has been
unveiled, sometimes dramatically, during his tenure,
I do think that elements of these objectives have been
consistent across multiple generations of Chinese
leadership. There are certain continuities
going back to Deng Xiaoping and his initial emphasis on science
(49:13):
and technology, on military modernization, on becoming
capable of more effective command of the PLA and shaping
it into a more modern fighting force.
So I do think there's been continuity over time, but
especially under XI Jinping's leadership, there has been this
focus on an innovation driven approach to economic development
(49:36):
and military modernization and this this strong emphasis on
technology as critical to reshaping the balance of power.
So I'd say that there's a feature that is a core element
of China's military, economic and overall national strategy is
(49:57):
betting on technologies that could be decisive to the future,
to the fourth industrial revolution, to this emerging
revolution. Military affairs.
And the desired end state does appear to be the PLA not just
being on par with the US military, but really being at
the forefront in a new era of warfare.
(50:19):
OK, brilliant wrap up. You nailed it.
Thank you. And I, I realised I forgot to
throw their cyber activity in there as one of the examples,
but that's entirely on me. So my own mission and we'll
we'll make it a subject for nexttime we talk, which hopefully
again we will be able to do. But Elsa, can you look?
Thank you so much for joining us.
That's hugely insightful into the way the PLA is modernising
it's military and it was a real pleasure having you on Stop the
(50:40):
World. I thank you so much and looking
forward to continuing the conversation again sometime.
Thanks for listening folks. We will be back next week with
another episode. Bye for now.
Stop the world. The PLA is definitely actively
experimenting. There's some claims that some
tools have been deployed. There's definitely been
demonstrations of swarming, claims of weapon systems with
different degrees of autonomy, claims that decision support
(00:21):
tools enabled by AI are being deployed, including at the
theatre command level. Welcome to STOP the World, the
SP Podcast. I'm Olivia Nelson.
And I'm David Rowe. Dave, I love today's guest, Elsa
Kenya. She is brilliant.
She's smarter than my phone. I'll also bet that she's smarter
than GPT 5, which is rumoured tobe coming out this month.
(00:44):
She's damn smart anyway, and we're grateful that she's
devoted her efforts to understanding the Chinese
military. We all know it's made huge
progress in recent decades, and in terms of quantity, it's
outmatched in the US in some areas.
But just how good is it, especially not having fought a
war in decades? Yeah, so that's really the
debate that military experts around the world are having, and
Elsa knows as much about it as anyone in the public realm,
(01:07):
especially on the ways the People's Liberation Army has
incorporated technology. Elsa is a PhD candidate in
Harvard University's Department of Government, where she's just
recently defended her dissertation, China's Command
Revolution. She is an adjunct senior fellow
with the Centre for a New American Securities Technology
and National Security Programme,and she was previously a
(01:30):
Fulbright specialist and non resident fellow here at ASPY.
When I spoke with her this week,she was recovering from running
a half marathon live. And I was wondering what exactly
I'm doing with my life. Me too, Dave.
So Elsa talks about how the PLA matches up against the US
military, the recent history of its improvements, the concepts
(01:51):
of informatization and intelligentization, the
centrality of AI to its modernisation, and the Chinese
view of the ethics of automatinglethal force.
So it's quite a mouthful, but you got it out live well.
Yeah. Thanks, Dave.
She also talks about the significance and purpose of
China's military rehearsals around Taiwan, its concept of
peace disease, which translates as its own lack of real world
(02:12):
fighting experience in China's overall strategy with its
growing military assertiveness in various arenas.
There's a lot packed into this conversation, but it rolls along
and really, Elsa is just a greatsource of insight into this
pretty murky area. So let's take a listen.
Welcome to STOP THE World, everyone.
I'm here with Elsa Kenya. Elsa, thanks for joining me.
(02:33):
Thanks so much, happy to be here.
So let's start by looking at a high level at the progress that
the PLA has made in recent years.
The US has deemed it a pacing threat for a while now.
I mean, we don't have to go through the whole history, but
as a way of giving an overview to the conversation perhaps, how
would you quickly characterise the state of the PLA?
(02:56):
Where has it made the most advances since, say, 2000?
Where is it still deficient? And just in headline terms, you
know, where is it strongest? Sure.
So the PLA has achieved dramaticprogress but continues to
confront considerable challengesand I'd I'd point to significant
(03:21):
expansion of Chinese cyberspace capabilities as well as space
and counter space. There's the PLA rocket force of
course and dramatic developmentsand hypersonics as well.
But I think some of the deficiencies may persist in the
lack of operational experience, questionable capabilities of
(03:42):
Chinese military commanders, thepotentially some improvement on
that front given attempts to upgrade their training and their
education and cultivation of talent.
So I think assessing Chinese military power is inherently
complex and uncertain. I tend to think that the truth
lies somewhere in the middle, between those who would dismiss
(04:05):
the PLA as a military museum andthose who would argue it's
already leagues ahead of the US military.
It's hard to right size our understanding of Chinese
military power, but I think it'simportant to both recognise the
progression but also understand that the PLA still sees itself
as striving to catch up with theUS military, which it regards as
(04:27):
a powerful adversary and the benchmark for its own
modernization. And there is still a level of
anxiety about the US military moving ahead into new
capabilities, pursuing new technologies.
There is a fear of falling behind or of technology
surprise. But at the same time, there does
(04:48):
appear to be a greater degree ofconfidence today that the PLA is
closer to closing that gap, especially when it comes to new
domains and emerging capabilities, whether that is
cyberspace or artificial intelligence or exciting
developments in quantum technology as well.
Depending on the time horizons, different capabilities may have
(05:10):
different degrees of relevance in in various contingencies.
But the PL as ambition is clear to not just pursue a strategy of
catching up, of developing asymmetric capabilities to
counterbalance the US military, but now looking as well to
achieve the leapfrog development, so to speak, and of
(05:32):
genuine disruptive innovation and trying to achieve an
advantage or the strategic commanding heights, as Chinese
military commentators often phrase it, in some of these new
domains and technologies where the US does not have a clear or
decisive advantage. So I I think we should be
sceptical sometimes of the claims in Chinese military media
and propaganda, but also take these competitive challenges
(05:55):
quite seriously. I mean, this is a layperson's
question, I suppose, but is it possible to put them side by
side and say the PLA is, I don'tknow, 70% of the way there, 80%
of the way there, something likethat?
I am hesitant to put a number onit in the sense that it's apples
and oranges in some sense for lack of a better metaphor.
That talking about two militaries with very different
(06:17):
histories, different ambitions and optimise for different
scenarios. Where U.S. military has been
focused on power projection and has recent experience in the
Middle East primarily which has varying degrees of relevance
contingencies in the Indo Pacific depending on one's
vantage point. Whereas the PLA does not have
(06:38):
that contemporary operational experience and is very focused
instead of on being a global military power, on regional
contingencies, specifically Taiwan, and sort of first on
counter intervention campaigns on preventing U.S. military
projection of power. So it's almost thesis and
antithesis in some sense where we're looking at different
(07:00):
paradigms of military power. Of course, we've also seen the
PLA try to emulate some of the hallmarks of U.S. military
power, as in the development of aircraft carriers.
But the PLA is focused much moreso on asymmetric capabilities
and on counter intervention and developing capabilities to
target the weaknesses it recognises in US paradigms of
(07:22):
military power. Yeah.
OK, Now that's fair enough. Some of your work in the work of
others has looked at how they'veput information at the heart of
their modernization. Can you just talk a little bit
about that? I mean, in what sense has their
focus on information really kindof, I suppose, shaped their
modernization? So I'd say that the PLA, since
at least the 1990s has really concentrated on information as a
(07:47):
central feature of military power, focusing on information
power as an enabler of combat capability and seeking to
achieve information dominance inthe early stages of any conflict
scenario. And I'd say that is critical to
how the PLA regards the former character of conflict.
(08:09):
And this focus on information has driven changes in Chinese
military strategy and their doctrine and sort of core
concepts at the heart of how thePLA is preparing to fight.
And we can look back at the lessons the PLA learned from the
Gulf War and from US conflicts in Iraq and Afghanistan and
(08:30):
looking at sort of network centric warfare and trying to
develop their own approaches focused on what they call
informationization. Xin XI Hua, which really is this
transformation of warfare and ofmilitary power through
information. And that has driven the
development of capabilities for information operations, namely
(08:52):
cyber, electronic and psychological warfare focused on
targeting their adversaries, decision making and and
targeting vulnerabilities in battle networks as well.
And then also developing greaterresiliency in their own C4 ISR
capabilities, their own command information systems and trying
to ensure a degree of resilienceand redundancy having seen how
(09:16):
the US and its conflicts has often targeted those sorts of
networks and systems. So this is also closely linked
to concepts of sort of system ofsystems confrontation as a core
feature of modern and future warfare.
And increasingly, beyond this emphasis on informatization,
(09:36):
there's also this focus on the impact of artificial
intelligence as a new catalyst for changes in the form of
warfare, which is leading to these these emerging and
evolving debates on intelligentization or the
transformation of warfare based on advances in AI and related
emerging technologies that are seen as an element of a
(10:00):
revolution in military affairs. And US debates on the revolution
in military affairs sort of dwindled somewhat after the
early 2000s, whereas the PLA hasconsistently seen this framing
of military revolution or transformation as really pivotal
to how they are approaching, initially catching up in
advances in information technology and today trying to
(10:22):
achieve a leading position and new frontiers of artificial
intelligence. Yeah, yeah.
And I definitely want to come back to AI and B
intelligentization. So I mean, from what you're
saying, I mean they're looking at it both from a point of view
of information warfare in terms of, of operations looking for
vulnerabilities in adversaries, I suppose information networks
in order to potentially disrupt those.
(10:44):
But also information, I suppose harmonisation within their own
forces to, you know, improve their ability to fight as a sort
of coordinated and joint force. If I, if I understand it
correctly, I'm interested in, you know, the third of those
categories, what it's meant for their command and control.
I mean, that was, I think their C2, as it's called, was
(11:05):
traditionally seen as a bit ropey.
They had difficulty with, you know, I suppose the the lack of
initiative at shown at more junior levels and the the
strategic goals set by the the higher levels making their way
down. And so basically their command
and control was well below wherethey wanted it to be.
What has this information utilisation, if that's the word?
(11:27):
What has that meant for their command and control in
particular, and how good is it now?
So I guess I'll start with some of the self criticisms you
referenced that dating back to at least 2015, Chinese military
leaders and commentators, including up to the level of Xi
Jinping and sort of senior members of the Central Military
(11:48):
Commission were quite candid in pointing out what they saw as
shortfalls in the commanders capabilities.
The so called 5 incapables, including that commanders at the
time were quote unquote. Some commanders had difficulty
understanding the intentions of leadership, of judging the
battlefield situation, of decision making and uncertain
(12:10):
situations and so on. And that this this actually was
the driver for new approaches totraining focused on trying to
overcome these self assessed shortcomings and to ensure that
commanders could be more capableof exercising their subjective
(12:30):
initiative and interpreting senior leadership guidance,
including understanding the party's intentions and being
able to respond to more complex and dynamic battlefield
situations. So I think when we hear these
self criticisms and often sort of a tendency to echo these in
in debates among external observers and how to right size
(12:51):
Chinese military power. I think it's important to note
that the PLAC has been very forthright in discussing these
problems and has been very targeted and focused in its
efforts to overcome some of these challenges in these human
elements. And certain more recent
(13:12):
commentaries still allude to thefive incapables and other self
criticisms as not fully resolved.
But I'd say in the past decade there has been a certain amount
of progress or some indicators that the more realistic
training, the effort, efforts toreform elements of their
(13:32):
military education and cultivation of joint commanders,
as well as grassroots commanderstrying to encourage them to be
more creative and adaptive are starting to gain some traction.
So I do think it's important to recognise that the PLA is a
learning organisation and they are adapting.
They are trying to overcome the weaknesses that they recognise
(13:53):
that relate to the peace diseasethat they see as a problem given
their lack of contemporary operational experience.
And there are reasons to be sceptical that they can fully
overcome these problems. But I think it's important not
to discount the possibility thatthey may achieve greater
progress than we may appreciate.Though of course, for those of
(14:14):
us who watch the PLA, based on PLA media and military writings,
there's always the caveat that so much of the information we
rely upon is generated from within the system.
So we see a level of candour, but there's also can be a
certain amount of distortion andobviously controls and
censorship and propaganda associated with some of this
(14:37):
representation as well. So I think it's important to be
to be sceptical, but also recognise that there the PLA has
invested and aligned its training to overcome some of
these previously identified deficiencies and commanders
capabilities. So that's the human element of
it. From a technical perspective,
(14:58):
there have been significant investments in the resilience
and redundancy of their whole architecture of command control
communications, as well as the information surveillance and
reconnaissance capabilities thatconstitute elements of the
information support that would enable precision strikes.
(15:19):
And the stand up of the Information Support Force in
April of 2024 based on an elevation of elements of the
former strategic support force, I think reflects the importance
and prioritisation associated with continuing to advance their
command information systems and the complex network information
systems that are recognised as critical capabilities for
(15:42):
current and future conflict. So say that there have been
major advances in China's deployment of space based
systems for surveillance to support strikes and targeting.
There's been very redundant, robust ground based architecture
(16:02):
of information networks to enable insured communications
and continued attempts to innovate and integrate new
technologies into this overall architecture of command and
network information systems, focusing on improving this
overall capacity for informationsupport and including greater
(16:23):
decision support. So there may be a tendency to
turn to technology to compensatefor some of these human
shortfalls. But I think inherently there's
also a pragmatic recognition that the complexity of
contemporary operations is such that no human commander, no
matter how capable, can make decisions effectively unless
(16:45):
they have the right information and the right analytic
capability and the right intelligence.
So I think information support and greater intelligentization,
so to speak, in the processing of ISR and use of some of these
new tools and techniques to support command decision making
is also critical. So this focus on improving the
(17:05):
human element and upgrading the technology are proceeding in
parallel. Yeah.
OK. Now, I was going to ask about, I
mean, that parallel dimension isa really interesting one.
I mean, the technology has improved, but at the same time,
I suppose their philosophy, their doctrine, what you will,
has shifted at the same time towards this approach of let's
make sure we can see what we need to see, share the
(17:28):
information we need to share in order to be able to make good
decisions, give and receive the orders that we need to give and
receive. And that's turned them from what
you're saying into into a very aware, very commutative and and
potentially very decisive force overall.
Yeah. Absolutely.
And I do think that there is this focus on integrating and
(17:50):
reconciling these human and machine or technological
dimensions of warfare. I've been asked a number of
times over the years, do I thinkthe PLA will take humans fully
out of the loop and have machineintelligence commanding all
operations. And I tend to be sceptical of
that. But I think there, I always say
I'm highly confident in the nearterm at least one human will be
(18:11):
in the loop for all strategic decisions.
That would be Xi Jinping personally, his role as
commander in chief. But when it does come to more
tactical and technical operations, there appears to be
a fairly flexible and pragmatic approach to leveraging emerging
technologies for applications they're uniquely suited for.
So for instance, greater automation of elements of
(18:34):
cyberspace operations, for instance, including uses of
generative artificial intelligence to accelerate and
enable scaling up of some of these capabilities and
exploitations I think is something we should expect to
see and some evidence there's experimentation along those
lines. Applying AI tools, including at
the edge to accelerate processing of intelligence,
(18:58):
including from some of their space systems is something we
can expect to see even some start UPS launching what they're
describing as AI enabled satellites.
So think a lot of very technologically deterministic or
solutionistic approaches to leveraging technology where it
has certain strengths. But there also appear to be
(19:19):
fairly nuanced debates within the PLA about the strengths and
weaknesses of machine intelligence relative to human
intelligence and the recognitionthat there can be certain biases
or issues of safety or surety ofthese systems, which
necessitates A continued human involvement and command decision
(19:39):
making at the operational strategic levels.
Even if certain tactical applications, elements of air
and missile defence, for instance, or aspects of
swarming, suicide, drones and and the like could be more fully
automated. So I think there's not a single
answer or a single approach in terms of how the PLA may use
(19:59):
these tools or may approach integrating technologies.
The lot of this will depend on different domains, applications
and the evolving maturity of these capabilities.
As you've sort of anticipated where I wanted to go next in a
wonderful way, I mean that that is a really just a fascinating
insight into the into the way they're adopting AI and
(20:20):
automation. And at this point, I just want
to give you a quick shout out for the 2019 Aspy report that
you wrote on Chinese Russian cooperation in areas like AI.
So fully six years ago, not justwere you anticipating the China
Russia cooperation, which of course has become one of the
biggest issues the strategists are considering in 2025, but
(20:40):
also the importance of things like AI and other advanced
technologies. So a quick shout out there, but
just, I mean, that's thrown up an ethical question for me.
I mean, it's one that's as old as the hills, but the basic idea
of using lethal force without a human in the has generally been
seen as a red line for militaries who are answerable to
democratic systems. Do you think that is not an
(21:02):
ethical red line for the PLA? So my impression would be that
the PLA is very pragmatic in their approach.
There are debates in Chinese military writings about ethics,
about law of armed conflict, about the benefits or drawbacks
(21:23):
of having a human in, on or out of the loop.
But with the caveat that these days discussing directly with
Chinese military counterparts ismore challenging, my impression
of these debates is that the PLA's approach is relatively
(21:45):
pragmatic. And I suppose I keep coming back
to that characterization becausethere does appear to be a degree
of concern with law of armed conflict, legitimacy and use of
force, a desire to have precise and accurate targeting as
opposed to great risk of collateral damage.
(22:07):
And the PLA is a professional military.
They're not looking to cause mass casualty events without of
control autonomous weapons. They want to ensure that their
weapon systems that may have various degrees of automation or
autonomy built in will hit the targets they intend for them to
hit. So I do think that there is a
(22:30):
active debate or there appears to be ongoing consideration of
the of the relative advantages and disadvantages of having
human decision makers in on or out of the loop at different
echelons for different instancesof employment of weapon systems.
But I think the notion that a human has to be pulling the
(22:53):
trigger, so to speak, metaphorically appears to be
less of a moral red line. And I think that there is.
Without weighing into these debates in US or Australian
context, I think there is something to be said for a
recognition that simply having ahuman quote, UN quote in the
loop is not enough for the decision to be an ethical 1.
(23:16):
I expect these will continue to be active debates in across
multiple militaries as the technology evolves, as the
calculus of risk evolves. And much may depend as well on
the operational environment. And the types of decisions or
capabilities that could be employed, for instance, in urban
(23:37):
warfare would differ greatly from the considerations if we're
thinking about naval warfare or maritime operations.
So I think the level of complexity and congestion of the
environment may be a factor. And I suspect the debates will
continue evolving in the PLA as here.
But the overwhelming considerations driving the PLA's
approach appear to be pragmatism, and the question of
(24:02):
which approach is technically and operationally advantageous,
as opposed to more abstract notions of red lines morally.
Yeah, yeah. No, you're right.
It's far from a simple discussion.
You wrote recently about the sort of jolt that China and and
including Xi Jinping personally got from the famous AI battle
(24:24):
between Alpha Go and the then Goworld Champion Lisa Doll and the
trajectory. I mean, famously Alpha for Go
defeated Lisa Dole, much to everybody's surprise.
And it was a huge moment really well across all of Asia, where
go was this ancient and much loved and much played game,
probably missed in much of the West other than in circles that
(24:45):
were paying a lot of attention to to AI.
But you described recently that that sort of, you know, set
China on a trajectory. It was a sort of AI hate the
term Sputnik moment, but it was a bit of a Sputnik moment for
China. Just talk us through that a
little bit in the trajectory that it set China on in terms
terms of, I suppose, building AIcapabilities for including for
(25:06):
military purposes? Sure.
So looking back at Alphago and looking at the Pele's reaction
at that time, there did appear to be a catalytic impact to some
extent in terms of the recognition that this was a
demonstration of how quickly advances in AI were progressing
(25:27):
and the fact that an AI system could defeat the world's top
human players in a game. That did centre on decision
making, stratagems and complex calculations.
And think the salience of the game of Gove to some extent may
have been a factor there. But also the notion that during
(25:48):
some of these matches with Lisa Dole and other players, Alpha Go
started to introduce moves or plays that humans hadn't
conceived of. And the totality of the history
of the game of Go. So I think that.
The famous move 37, I think, wasn't it where, where everyone
was so aghast and and everyone thought that that the computer
had made it a terrible, disastrous mistake.
(26:09):
And it turned out to be a a brilliant strategy that, like
you say, no one had ever used before in the in the human
playing of the game. Indeed, and as someone who was
watching PLA debates and writings at that time and
initially having looked back, and this is history now, but the
PL as responses to the third offset strategy that was
launched around 2014 and through2015 when I was initially
(26:33):
started tracking these issues. The PLA both saw the strategic
and operational potential of these developments in the
context of their concerns about U.S. defence innovation
initiatives and believed there could be unique utility to
integrating AI into future command decision making.
(26:56):
And this prompted some of the early debates that had harkened
back to the PL as earlier interest in thinking about how
technology could enable an upgrade command decision making
on command as a complex system with human and machine elements
(27:16):
integrated. And this this strive to have a
more scientific approach to warfare and to command decision
making. And Alphago, in that context,
did spark a lot of debate and some apparent initial efforts
and experimentation with integrating AI and developing
tools for decision support. There's also a lot of efforts in
(27:38):
war gaming coming out of institutions like the Chinese
Academy of Sciences, the Palais National Defence University,
National University of Defence Technology.
A lot of focus on trying to develop AI agents that could
play war games and both AI versus human and AI versus AI
(27:59):
machine versus machine confrontation as means of
testing and validating how effectively AI tools or agents
could be used in military decision making, at least in a
games play and context. Of course there's a great
difference between a even the most sophisticated of
computerised war games relative to the complexity of an actual
(28:20):
real world operational environment.
But this was at least seen as a stepping stone or proof of
concept to validate how some of these AI agents or systems could
start to be developed and how they could be paired with human
players and that Centaur type approach to decision making as
well. So I do think that this was one
(28:42):
of several moments in the PL as recent history that did draw
their attention to the potentialof AI.
And I do find it ironic that this this moment was not seen
very much as of military relevance in AUS or generally
western context, but did spark acertain amount of debate in the
PLA, as did more recently their reaction to Chechnya PT and
(29:06):
concerned that at a moment whereChinese leaders and observers
have started to be more confident, they were closer to
drawing even with USAI developments.
ChatGPT, I think was another another moment of shock again,
I'll, I also hesitate to say a Sputnik moment, but certainly a
moment that had some catalytic impacts in driving efforts to
(29:27):
think about how large language models and generative AI
generally could contribute to future military capabilities,
whether in in intelligence and command decision making or in
other aspects of planning and strategizing.
So there is a a degree of reactivity and how the PLA has
(29:47):
responded to global developmentsin AI and how closely they've
monitored advances, especially in in AUS context, concern
always about the US as a powerful potential adversary and
fear of this fear of technology surprise.
That's a consistent theme. But there's also efforts to
beyond scanning global military and technological developments
(30:10):
to try to anticipate where some of these trends are going.
And not just to react and be a fast follower, but to be more of
a leader in original innovation and to design future warfare or
or design the parameters for thefuture battlefield.
So the PLA is I think a ways to go and truly achieving that.
But there's certainly an ambition and very close study of
(30:33):
technology and of some of the theoretical implications as
well. Yeah, well, we'll try not to
blame DeepMind and Open AI too much for catalysing China's
military modernization efforts. Is there a way to, in an apples
to apples sense, say who is doing a better job between the
PLA and the US military at actually integrating AI,
(30:58):
including in a human machine teaming sense?
That is a fantastic question andI don't have a great answer.
I think it is if. Anyone has it you do.
Well, it is an aspect of ongoingresearch for me.
I look more at the PLA side of the equation, so I'm less
confident making a comparison since I'm less fully familiar
(31:22):
with the US context here. But I think, I guess I'd say by
way of caveat, I think for both militaries, we're still at a
stage where there is a lot of experimentation.
There are a lot of efforts to have proofs of concept to try to
test and validate what systems or use cases and applications
(31:42):
can be feasible and advantageous.
There's a lot of very diverse efforts across every application
imaginable. Just about.
And at the risk of sounding cynical or flippant, I think
that sort of just at AI is, I think, a tendency sometimes of
military leaders in multiple systems to sort of have this
level of enthusiasm or exuberantexpectations of these
(32:05):
technologies that sometimes start to confront the realities
that when you are trying to implement complex advanced
technologies within military bureaucracies in any cultural
context, there will be obstacles.
And there will be difficulties with how you manage data, how
you upskill your personnel at different echelons to use these
tools effectively. How you start to develop new
(32:29):
procedures and parameters for integrating these technologies
into operations, including in training and exercises and
recognising some of the problemsthat arise.
So I guess I'd say ask me again in a couple years.
I think the PLA is definitely actively experimenting.
There's some claims that some tools have been deployed.
There's definitely been demonstrations of swarming
(32:52):
claims of weapon systems with different degrees of autonomy,
claims that decision support tools enabled by AI are being
deployed, including at the theatre command level.
Lot of applications. I think some of the earliest
applications we'll see will be in virtual domains where it's
harder to measure or observe from the outside, whether that
(33:13):
is in accelerated processing of data to enable enhanced
intelligence capabilities or integrating generative AI for
cognitive domain or cyberspace operations.
So I think some of these earliest manifestations may be
in virtual domains or in some ofthese information related
(33:35):
functions and capabilities. There's been some some
indications of AI adoption in different aspects of cyberspace
operations and that could including for malware
development, social engineering and the like generation of
content for more targeted psychological operations.
So I think think some of the near term applications will
(33:59):
observe in competition or confrontation will be along
those lines how effectively certain capabilities are being
integrated into operational decision making or harder to
observe from the outside. So it's easier to point to the
indications of effort and progress, harder to say how
effective or successful these initiatives are.
(34:21):
And I guess I'd add as well thatthere there is I think a greater
extent of uncertainty about the military balance because these
qualitative advancements could have significant applications
and implications for the future of warfare.
But there is an inherent difficulty in measurement.
It's hard to count or hard to measure how much having a
(34:43):
certain AI modular system deployed enhances targeting if
it's say an automatic target recognition and Chinese
missiles. It's hard to hard to measure or
count or evaluate in the abstract.
And I think this also highlightswhy we need greater integration
of subject matter and technical expertise.
(35:03):
I'll be the first to admit I'm as political scientists by
training, I am always trying to continue learning and upscaling
myself. I'm actually here at Berkeley
for an AI red team in boot camp this week, which has been a
fantastic experience so far. So definitely a proponent of
lifelong learning for all of us to keep pace with this.
But I do think that for those ofus who've been observers of
(35:23):
military affairs, I think the how we calculate or how military
strategists and decision makers calculate the future military
balance may depend more and moreon complex technologies for
which the metrics are difficult and uncertain.
Which I think does also increasethe risk of some kind of
miscalculation or misperceptionsthat could arise with that new
(35:47):
fog of technology, so to speak. I'm envious of where you are now
and I'm just picking up on your point.
I think that, you know, the challenges that you were
referring to there of, you know,just add AI and assuming that
integration and adoption is straightforward is a challenge
that in fact large organisations, military, private
sector, civilian of all kinds finding.
(36:07):
And it's, it's fascinating that one of the, the bottlenecks, as
it turns out, is, is simply organisational readjustment in
order to, to integrate AI. It's interesting, but unless I
go down a rabbit hole of my own digging there, let's just there
are a couple of other things that I do desperately want to
cover off on while we're here. And what 1 is just you've
touched on the PL as lack of, you know, real World War
(36:30):
fighting experience and you've sort of indicated that that is
being overcome in some ways. How relevant or significant does
that remain? And to what extent do you think
that the rehearsals around Taiwan as Admiral Paparo Indo
Picom commander has described them?
To what extent are those actually about, you know,
(36:51):
practical experience and and learning and actually improving
in that real world scenario? The PLA is quite candid in
admitting to the challenges it'slack of contemporary operational
experience can present. The so called peace disease
which is often referenced in commentaries which are often
accompanied by expectations for more combat realistic or actual
(37:16):
combat training. And the extent to which the PLA,
especially the Eastern Theatre Command and its continual
presence and operations around Taiwan can get experience in the
likely operational environment for its priority scenario and
can validate command relations, complex multi domain
(37:39):
synchronisation in some of thesepressure operations.
And that is perhaps at least a partial mechanism of overcoming
that. So this focus on greater
sophistication in training on the integration between
peacetime employment of militaryforces with likely wartime
(38:03):
contingencies that all is aimed at overcoming some of these
factors, as are the emphasis on war gaming simulations and this
fairly scientific approach to trying to anticipate some of the
challenges that could arise in afuture conflict scenario.
(38:25):
Think military observers have been quite wrong in, for
instance, assessments of Russiancapabilities.
So somebody who studies the PLAItend to be quite humble and
cognizant of the limitations in in our community's perspective
and what we see and don't see. But I do think the fact that the
(38:46):
PLA, especially the Eastern Theatre Command, which has been
responsible for orchestrating these complex pressure
operations, has been seemingly successful, at least as
portrayed in their own media andundertaking these named
exercises and major campaigns around Taiwan.
(39:07):
That is seemingly advantageous in their familiarisation, in
their capacity to rehearse as Admiral Papparo's characterised
it, and as contributes to overall readiness.
So I definitely don't think we should dismiss peace disease,
and certainly the PLA does not dismiss this as an impediment,
(39:27):
but so too do our our own militaries not have experiences
that quite align with the complexity of what a future
conflict in this theatre could entail.
I hadn't heard the term peace disease before.
It's a bloody depressing phrase to hear.
Indeed. You you make a good point about
(39:48):
the Russian military there as well, and how it reflects on the
humility we should show in making bold predictions about
any fighting force. OK, are you prepared to say how
red do you think the PLA is to mount a successful blockade or
even invasion of Taiwan at the moment?
So I would say the current posture as apparent and well
(40:13):
documented of the PLA in the vicinity of Taiwan gives them
the optionality to transition into a blockade or quarantine
should they choose to do so. And the strategic calculus or
rationale is something of a question.
But in terms of the the extent to which the PLA has expanded
(40:37):
and attempted to normalise theirposture around Taiwan is
significant and has expanded dramatically, especially since
Speaker Pelosi's visit, which the PLA chose to respond to with
massive pressure operations thatthey justified as a proportional
(40:59):
reaction, but did create a sea change and a step towards
expanding their posture in ways that give them that optionality
and capability to transition into a blockade.
And certainly that has been someof the signalling not not so
(41:19):
subtle in some cases associated with elements of their joint
Sword or Straight Thunder exercise series.
So a blockade or quarantine and there there would almost
certainly be some kind of legal warfare or political warfare
justification accompanying it asan attempt to rationalise it
(41:41):
that would not be all that operationally challenging based
on how they are already postured.
And I think this incremental expansion in attempts to change
facts on the water has been significant.
But that is a far cry from readiness for an actual
invasion. And I do tend to be of the
(42:03):
school of thought that although the threat of an invasion is not
trivial and shouldn't be dismissed, there are many tools
and techniques through which thePLA could escalate their
posture, presence and pressure around Taiwan in an attempt to
(42:27):
deter or intimidate or attempt to force acquiescence prior to
attempting an invasion. Because an invasion would be
incredibly challenging. So the PLA has been evidently
directed by Xi Jinping to be prepared by 2027.
Which again, I think often this point gets conflated in some of
(42:52):
the media coverage and public debates.
But becoming capable by 2027 does not imply an intention or a
deadline or timeline centre in 2027.
But the the benchmark has seemingly been established as
their centenary objective of becoming a powerful military on
(43:13):
a course to becoming a world class force and assuring China's
national rejuvenation by mid century and so on and so forth.
That they are seeking to have the capability to invade,
including as a means of attemptsto coerce Taiwan.
But I think I would suspect thatthe PLA today is quite a ways
(43:36):
from actually becoming confidentit has the ability to do so or a
decision to invade would be a major gamble on the part of
Chinese leadership and there would be non trivial risks in
that choice. So I do think that there are
reasons for concern. We could see orthogonal
(44:00):
approaches or horizontal escalation or different
techniques trying to target Taiwan and also trying to target
US calculus, since I think ultimately Chinese leaders and
observers appear to regard US decisions and U.S. military
(44:21):
power as the primary, if not predominant obstacle to their
capacity to achieve unification as they term it.
So I think we can expect to see more creative techniques and
attempts to impact US calculus or Taiwan's decision making
(44:43):
prior to any actual attempted invasion, but that does not mean
that the threat is not quite serious.
Yeah, yeah. Fascinating.
All right. And now, Elsa, to wrap up, I'm
going to be a bit grossly unfairhere, but I think you're so good
on this stuff that you can do it.
But I'm going to ask you to like, in roughly a minute or so,
(45:03):
taking account of China's military modernisation, its
activity around Taiwan, it's activity in the South China Sea,
the East China Sea, you know, construction of bases further
afield globally. How would you quickly describe
(45:23):
Beijing's overall military strategy in the service of its
broader national interest? So I'd say that when it comes to
activities within the first island chain, East and South
China Seas, Taiwan Strait, Chinese leaders are trying to
(45:46):
transform China into a major andgreat maritime power that has a
dominant position within its near seas, within the first
island chain and can exercise effective control of those
waters. I do.
I believe and I'm concerned thatChinese leaders, including Xi
(46:09):
Jinping especially, regard unification with or de facto
coercive annexation of, if necessary, of Taiwan by mid
century as among the benchmarks for national rejuvenation or the
XI Jinping's speeches and statements have articulated that
(46:32):
objective. And I do think that Chinese
military strategy and capabilities development are
aimed at counter intervention, and namely countering what
Chinese leaders regard as the potential for US intervention
(46:54):
that is seen as an obstacle to achieving these objectives, as
an animating driver of this asymmetric counterbalancing
strategy, as some Chinese military textbooks describe it.
So I do think that there are efforts to achieve incremental
advancement towards these aims. Xi Jinping and some of his
(47:18):
statements and the growing number of commentaries highlight
peacetime employment of militaryforces as an important element
of this approach of as the PLA becomes a more capable military,
the ability to brandish that power and to achieve deterrent
or coercive effects on China's neighbours in ways that
(47:39):
strengthen their position. And some of these territorial
disputes in the region is is emerging as a major feature of
their approach as well. And I do think that ultimately
this is driving to this ambitionby mid century to be a world
class or global first class military, which appears to imply
not only relative dominance in the region, but a transformation
(48:05):
of the military balance. The Chinese strategists and
observers appear to be becoming increasingly confident they are
on track to achieve and that notonly this transformation of the
cross Strait balance, but also adramatic alteration of the US
China military balance that is often accompanied by or often
(48:28):
accompanies some of these claimsthat the overall balance of
power, of comprehensive nationalpower is shifting in China's
favour. I believe there is a degree of
overconfidence in some cases that is juxtaposed with intense
insecurities on on other fronts.But the ambition is clear.
(48:50):
And although Xi Jinping gets a lot of credit for what has been
unveiled, sometimes dramatically, during his tenure,
I do think that elements of these objectives have been
consistent across multiple generations of Chinese
leadership. There are certain continuities
going back to Deng Xiaoping and his initial emphasis on science
(49:13):
and technology, on military modernization, on becoming
capable of more effective command of the PLA and shaping
it into a more modern fighting force.
So I do think there's been continuity over time, but
especially under XI Jinping's leadership, there has been this
focus on an innovation driven approach to economic development
(49:36):
and military modernization and this this strong emphasis on
technology as critical to reshaping the balance of power.
So I'd say that there's a feature that is a core element
of China's military, economic and overall national strategy is
(49:57):
betting on technologies that could be decisive to the future,
to the fourth industrial revolution, to this emerging
revolution. Military affairs.
And the desired end state does appear to be the PLA not just
being on par with the US military, but really being at
the forefront in a new era of warfare.
(50:19):
OK, brilliant wrap up. You nailed it.
Thank you. And I, I realised I forgot to
throw their cyber activity in there as one of the examples,
but that's entirely on me. So my own mission and we'll
we'll make it a subject for nexttime we talk, which hopefully
again we will be able to do. But Elsa, can you look?
Thank you so much for joining us.
That's hugely insightful into the way the PLA is modernising
it's military and it was a real pleasure having you on Stop the
(50:40):
World. I thank you so much and looking
forward to continuing the conversation again sometime.
Thanks for listening folks. We will be back next week with
another episode. Bye for now.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.