June 26, 2025 • 11 mins
In this episode, Bob introduces DeepMind's AlphaGenome, exploring its potential for revolutionizing genomics with AI. The discussion moves to Google's launch of the Gemini Command Line Interface, highlighting its new capabilities for developers. Bob then examines a U.S. legislative proposal aiming to ban Chinese AI systems from federal agencies, considering its implications for national security and international relations. The episode continues with an introduction to Metaview's innovative recruitment technology, which promises to disrupt traditional hiring processes. The episode concludes with closing remarks and a reminder about subscription options.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
What if we could predict how genes areregulated just by looking at their DNA
(00:04):
sequence?
Welcome to The AI News Daily Brief, your go-tofor the latest artificial intelligence updates.
Today is Wednesday, June 25th, 2025.
Here’s what you need to know about DeepMind'sgroundbreaking new model, AlphaGenome.
Let’s dive in.
(00:26):
DeepMind, Google's artificial intelligenceresearch company, is making waves in the world
of genomics with their latest innovation,AlphaGenome.
This artificial intelligence model is designedto predict gene expression from DNA sequences.
It is a fascinating leap forward, especiallyfor non-commercial researchers who can use it
(00:47):
to examine candidate mutations, testhypotheses, and design experiments.
DeepMind has already made a name for itself inthe scientific community with its deep learning
model AlphaFold, which predicts proteinstructures and even shared a Nobel Prize in
chemistry last year.
But now, they're tackling the more complexchallenge of understanding how DNA encodes gene
(01:11):
regulation.
According to DeepMind research engineer NatashaLatysheva, genomics is a 'fuzzy field' without
a single metric of success, so the team aimedto address as many aspects as possible with
AlphaGenome.
This development could be a game changer intherapeutic development, providing insights
that could lead to new treatments and medicalbreakthroughs.
(01:34):
Imagine the potential of being able to predicthow genes will behave and interact just by
analyzing their DNA sequences.
It's a thrilling time for both the artificialintelligence and genomics fields, as this
technology could pave the way for unprecedentedadvancements in healthcare and biology.
Google is making a bold move to bring itsartificial intelligence models closer to the
(01:58):
fingertips of developers with the launch ofGemini Command Line Interface, a new
open-source AI tool.
This tool is designed to integrate seamlesslywith developers' coding environments, providing
a powerful ally right in their terminal window.
Imagine having the ability to ask your codingtool to explain a convoluted section of code or
(02:19):
even to help write new features and debugexisting ones.
That’s exactly what Gemini Command LineInterface promises to do.
It connects Google's Gemini artificialintelligence models directly to local
codebases, making it easier for developers toincorporate artificial intelligence into their
everyday workflows.
(02:40):
The launch of Gemini Command Line Interface ispart of Google’s broader strategy to encourage
developers to use its artificial intelligencemodels in their coding processes.
They’ve already rolled out other AI codingtools like Gemini Code Assist and the
asynchronous AI coding assistant, Jules.
But with Gemini Command Line Interface, Googleis stepping into direct competition with other
(03:02):
command-line AI tools like OpenAI’s CodexCommand Line Interface and Anthropic’s Claude
Code.
Since the release of Gemini 2.5 Pro earlierthis year, Google’s AI models have become
increasingly popular among developers.
This popularity has also boosted the use ofthird-party AI coding tools like Cursor and
(03:25):
GitHub Copilot, which have grown into massivebusinesses.
In response, Google is keen to establish adirect relationship with developers by offering
in-house products.
But here’s where it gets even more interesting.
While most developers will likely use GeminiCommand Line Interface primarily for coding
tasks, Google has designed it to handle avariety of other functions as well.
(03:50):
Developers can use it to create videos usingGoogle’s Veo 3 model, generate research reports
with the Deep Research agent, or even accessreal-time information through Google Search.
Plus, it can connect to MCP servers, allowingfor external database connections.
To sweeten the deal, Google is open sourcingGemini Command Line Interface under the Apache
(04:14):
2.0 license, one of the most permissivelicenses available.
This move is expected to foster a community ofdevelopers contributing to the project on
GitHub.
They're also offering generous usage limits toencourage adoption, allowing free users up to
60 model requests per minute and 1,000 requestsper day, which is double what developers
(04:36):
typically used before.
Of course, as with any AI tool, there are risksinvolved.
A 2024 survey from Stack Overflow found thatonly 43 percent of developers trust the
accuracy of AI tools.
Studies have shown that code-generating AImodels can occasionally introduce errors or
miss security vulnerabilities.
(04:57):
So while the tools are powerful, developersshould remain vigilant and double-check the
output.
There's a new bill making waves in Washington,aiming to ban Chinese artificial intelligence
systems from federal agencies.
This move comes as part of the United States'strategy to secure its position in the global
artificial intelligence race against China.
(05:19):
Lawmakers are not just talking about it—they'retaking action to ensure the United States
remains at the forefront of artificialintelligence development.
Representative John Moolenaar, the Republicanchair of the House Select Committee on China,
did not mince words.
He described the situation as a 'new Cold War,'with artificial intelligence being the
(05:41):
strategic technology at its core.
According to him, the future balance of powercould very well hinge on who leads in
artificial intelligence.
Now, the concern is real.
Just a few months ago, a Chinese start-upcalled DeekSeek introduced an artificial
intelligence model that rivaled those fromindustry giants like OpenAI and Google, but at
(06:03):
a fraction of the cost.
This raised alarms that China might be closingthe gap, despite existing restrictions on
crucial technologies like chips.
The stakes are incredibly high, and the U.S.
must win, say experts like Thomas Mahnken fromthe Center for Strategic and Budgetary
Assessments.
This long-term competition is not just abouttechnology—it's about shaping the global
(06:26):
political order for years to come.
Jack Clark from Anthropic highlighted a crucialpoint: artificial intelligence systems reflect
the values of the societies they come from.
He argues that artificial intelligencedeveloped in democracies will lead to better
technology for humanity, while artificialintelligence from authoritarian regimes could
(06:47):
be imbued with those same authoritariancharacteristics.
The No Adversarial AI Act is already on thetable, introduced by both Republican and
Democratic lawmakers.
This bill seeks to identify and ban artificialintelligence systems from foreign adversaries
in the U.S.
government, with some exceptions for researchand counterterrorism.
(07:09):
It’s a decisive step in drawing a line againsthostile artificial intelligence infiltration.
And there’s more.
The U.S.
is strengthening export controls on advancedchips to China, a move that’s seen as essential
in keeping the upper hand.
As Jack Clark put it, the competition'fundamentally runs on compute.' Without
(07:30):
controlling the flow of these powerful chips,the U.S.
risks giving China the tools needed to developartificial intelligence that could threaten
American interests.
This is not just about technology—it's aboutnational security and global influence.
The AI Index Report by Stanford Universityshows the U.S.
(07:50):
still leads in producing top artificialintelligence models, but China is catching up
fast.
They're leading in artificial intelligencepublications and patents, signaling an intense
race that’s far from over.
Imagine a world where the hiring process is asseamless as booking an Uber ride.
That's what Metaview, a new artificialintelligence startup, is aiming to achieve.
(08:14):
Welcome back to The AI News Daily Brief, wherewe dive into the latest advancements in
artificial intelligence.
I am Bob, your host, and today we’re exploringhow former Uber and Palantir executives are
shaking up the recruitment industry with athirty-five million dollar investment to
disrupt corporate hiring.
Metaview has just raised thirty-five milliondollars in a Series B funding round led by
(08:38):
Google Ventures.
Founded in 2018 by Uber and Palantir alumsSiadhal Magos and Shahriar Tajbakhsh, this
startup is on a mission to transform the waycompanies hire by leveraging the power of
artificial intelligence.
Their flagship product is an AI note-taker thatrecords and structures interview notes, freeing
(08:58):
hiring managers from the tedious task of manualnote-taking and allowing them to focus on what
really matters—finding the right candidates.
But that is just the beginning.
Metaview plans to build a comprehensive suiteof artificial intelligence tools designed to
streamline the entire hiring process.
From generating job descriptions to optimizinghiring workflows, their tools aim to make
(09:23):
recruitment more efficient and data-driven.
Companies like Sony and Deliveroo are alreadyon board, and with this latest investment,
Metaview is set to expand its platform andteam.
The company's co-founder, Siadhal Magos, sawfirsthand the inefficiencies in hiring during
his time at Uber.
He realized that even at top-tier companies,the hiring process often relied on subjective
(09:48):
impressions rather than clear data.
This inspired him to create tools that harnessconversational data to improve decision-making
in recruitment.
As Magos puts it, 'AI is going to change how wework,' and Metaview is at the forefront of this
change.
The thirty-five million dollar investment willbe used to develop Metaview's full suite of
(10:10):
artificial intelligence tools, hire more staffat its London headquarters, and expand its
presence in San Francisco.
Google Ventures' Vidu Shanmugarajah believesthat recruitment is an area where
digitalization skipped a step, and Metaview isperfectly positioned to fill that gap with
cutting-edge artificial intelligence solutions.
(10:31):
While there are always concerns about theextent to which artificial intelligence should
be involved in hiring, especially with theEuropean Union’s classification of HR-related
artificial intelligence applications as 'highrisk,' Metaview is taking a careful approach.
The company insists that its artificialintelligence tools are designed to support, not
(10:53):
replace, human decision-making.
They aim to automate the administrative aspectsof hiring, leaving the critical human-to-human
interactions to people.
That’s it for today’s AI News Daily Brief.
With Metaview's bold move to revolutionizerecruitment, we're seeing how artificial
intelligence can streamline processes andenhance decision-making.
(11:15):
Thanks for tuning in—subscribe to stay updated.
This is Bob, signing off.
Until next time.
What if we could predict how genes areregulated just by looking at their DNA
(00:04):
sequence?
Welcome to The AI News Daily Brief, your go-tofor the latest artificial intelligence updates.
Today is Wednesday, June 25th, 2025.
Here’s what you need to know about DeepMind'sgroundbreaking new model, AlphaGenome.
Let’s dive in.
(00:26):
DeepMind, Google's artificial intelligenceresearch company, is making waves in the world
of genomics with their latest innovation,AlphaGenome.
This artificial intelligence model is designedto predict gene expression from DNA sequences.
It is a fascinating leap forward, especiallyfor non-commercial researchers who can use it
(00:47):
to examine candidate mutations, testhypotheses, and design experiments.
DeepMind has already made a name for itself inthe scientific community with its deep learning
model AlphaFold, which predicts proteinstructures and even shared a Nobel Prize in
chemistry last year.
But now, they're tackling the more complexchallenge of understanding how DNA encodes gene
(01:11):
regulation.
According to DeepMind research engineer NatashaLatysheva, genomics is a 'fuzzy field' without
a single metric of success, so the team aimedto address as many aspects as possible with
AlphaGenome.
This development could be a game changer intherapeutic development, providing insights
that could lead to new treatments and medicalbreakthroughs.
(01:34):
Imagine the potential of being able to predicthow genes will behave and interact just by
analyzing their DNA sequences.
It's a thrilling time for both the artificialintelligence and genomics fields, as this
technology could pave the way for unprecedentedadvancements in healthcare and biology.
Google is making a bold move to bring itsartificial intelligence models closer to the
(01:58):
fingertips of developers with the launch ofGemini Command Line Interface, a new
open-source AI tool.
This tool is designed to integrate seamlesslywith developers' coding environments, providing
a powerful ally right in their terminal window.
Imagine having the ability to ask your codingtool to explain a convoluted section of code or
(02:19):
even to help write new features and debugexisting ones.
That’s exactly what Gemini Command LineInterface promises to do.
It connects Google's Gemini artificialintelligence models directly to local
codebases, making it easier for developers toincorporate artificial intelligence into their
everyday workflows.
(02:40):
The launch of Gemini Command Line Interface ispart of Google’s broader strategy to encourage
developers to use its artificial intelligencemodels in their coding processes.
They’ve already rolled out other AI codingtools like Gemini Code Assist and the
asynchronous AI coding assistant, Jules.
But with Gemini Command Line Interface, Googleis stepping into direct competition with other
(03:02):
command-line AI tools like OpenAI’s CodexCommand Line Interface and Anthropic’s Claude
Code.
Since the release of Gemini 2.5 Pro earlierthis year, Google’s AI models have become
increasingly popular among developers.
This popularity has also boosted the use ofthird-party AI coding tools like Cursor and
(03:25):
GitHub Copilot, which have grown into massivebusinesses.
In response, Google is keen to establish adirect relationship with developers by offering
in-house products.
But here’s where it gets even more interesting.
While most developers will likely use GeminiCommand Line Interface primarily for coding
tasks, Google has designed it to handle avariety of other functions as well.
(03:50):
Developers can use it to create videos usingGoogle’s Veo 3 model, generate research reports
with the Deep Research agent, or even accessreal-time information through Google Search.
Plus, it can connect to MCP servers, allowingfor external database connections.
To sweeten the deal, Google is open sourcingGemini Command Line Interface under the Apache
(04:14):
2.0 license, one of the most permissivelicenses available.
This move is expected to foster a community ofdevelopers contributing to the project on
GitHub.
They're also offering generous usage limits toencourage adoption, allowing free users up to
60 model requests per minute and 1,000 requestsper day, which is double what developers
(04:36):
typically used before.
Of course, as with any AI tool, there are risksinvolved.
A 2024 survey from Stack Overflow found thatonly 43 percent of developers trust the
accuracy of AI tools.
Studies have shown that code-generating AImodels can occasionally introduce errors or
miss security vulnerabilities.
(04:57):
So while the tools are powerful, developersshould remain vigilant and double-check the
output.
There's a new bill making waves in Washington,aiming to ban Chinese artificial intelligence
systems from federal agencies.
This move comes as part of the United States'strategy to secure its position in the global
artificial intelligence race against China.
(05:19):
Lawmakers are not just talking about it—they'retaking action to ensure the United States
remains at the forefront of artificialintelligence development.
Representative John Moolenaar, the Republicanchair of the House Select Committee on China,
did not mince words.
He described the situation as a 'new Cold War,'with artificial intelligence being the
(05:41):
strategic technology at its core.
According to him, the future balance of powercould very well hinge on who leads in
artificial intelligence.
Now, the concern is real.
Just a few months ago, a Chinese start-upcalled DeekSeek introduced an artificial
intelligence model that rivaled those fromindustry giants like OpenAI and Google, but at
(06:03):
a fraction of the cost.
This raised alarms that China might be closingthe gap, despite existing restrictions on
crucial technologies like chips.
The stakes are incredibly high, and the U.S.
must win, say experts like Thomas Mahnken fromthe Center for Strategic and Budgetary
Assessments.
This long-term competition is not just abouttechnology—it's about shaping the global
(06:26):
political order for years to come.
Jack Clark from Anthropic highlighted a crucialpoint: artificial intelligence systems reflect
the values of the societies they come from.
He argues that artificial intelligencedeveloped in democracies will lead to better
technology for humanity, while artificialintelligence from authoritarian regimes could
(06:47):
be imbued with those same authoritariancharacteristics.
The No Adversarial AI Act is already on thetable, introduced by both Republican and
Democratic lawmakers.
This bill seeks to identify and ban artificialintelligence systems from foreign adversaries
in the U.S.
government, with some exceptions for researchand counterterrorism.
(07:09):
It’s a decisive step in drawing a line againsthostile artificial intelligence infiltration.
And there’s more.
The U.S.
is strengthening export controls on advancedchips to China, a move that’s seen as essential
in keeping the upper hand.
As Jack Clark put it, the competition'fundamentally runs on compute.' Without
(07:30):
controlling the flow of these powerful chips,the U.S.
risks giving China the tools needed to developartificial intelligence that could threaten
American interests.
This is not just about technology—it's aboutnational security and global influence.
The AI Index Report by Stanford Universityshows the U.S.
(07:50):
still leads in producing top artificialintelligence models, but China is catching up
fast.
They're leading in artificial intelligencepublications and patents, signaling an intense
race that’s far from over.
Imagine a world where the hiring process is asseamless as booking an Uber ride.
That's what Metaview, a new artificialintelligence startup, is aiming to achieve.
(08:14):
Welcome back to The AI News Daily Brief, wherewe dive into the latest advancements in
artificial intelligence.
I am Bob, your host, and today we’re exploringhow former Uber and Palantir executives are
shaking up the recruitment industry with athirty-five million dollar investment to
disrupt corporate hiring.
Metaview has just raised thirty-five milliondollars in a Series B funding round led by
(08:38):
Google Ventures.
Founded in 2018 by Uber and Palantir alumsSiadhal Magos and Shahriar Tajbakhsh, this
startup is on a mission to transform the waycompanies hire by leveraging the power of
artificial intelligence.
Their flagship product is an AI note-taker thatrecords and structures interview notes, freeing
(08:58):
hiring managers from the tedious task of manualnote-taking and allowing them to focus on what
really matters—finding the right candidates.
But that is just the beginning.
Metaview plans to build a comprehensive suiteof artificial intelligence tools designed to
streamline the entire hiring process.
From generating job descriptions to optimizinghiring workflows, their tools aim to make
(09:23):
recruitment more efficient and data-driven.
Companies like Sony and Deliveroo are alreadyon board, and with this latest investment,
Metaview is set to expand its platform andteam.
The company's co-founder, Siadhal Magos, sawfirsthand the inefficiencies in hiring during
his time at Uber.
He realized that even at top-tier companies,the hiring process often relied on subjective
(09:48):
impressions rather than clear data.
This inspired him to create tools that harnessconversational data to improve decision-making
in recruitment.
As Magos puts it, 'AI is going to change how wework,' and Metaview is at the forefront of this
change.
The thirty-five million dollar investment willbe used to develop Metaview's full suite of
(10:10):
artificial intelligence tools, hire more staffat its London headquarters, and expand its
presence in San Francisco.
Google Ventures' Vidu Shanmugarajah believesthat recruitment is an area where
digitalization skipped a step, and Metaview isperfectly positioned to fill that gap with
cutting-edge artificial intelligence solutions.
(10:31):
While there are always concerns about theextent to which artificial intelligence should
be involved in hiring, especially with theEuropean Union’s classification of HR-related
artificial intelligence applications as 'highrisk,' Metaview is taking a careful approach.
The company insists that its artificialintelligence tools are designed to support, not
(10:53):
replace, human decision-making.
They aim to automate the administrative aspectsof hiring, leaving the critical human-to-human
interactions to people.
That’s it for today’s AI News Daily Brief.
With Metaview's bold move to revolutionizerecruitment, we're seeing how artificial
intelligence can streamline processes andenhance decision-making.
(11:15):
Thanks for tuning in—subscribe to stay updated.
This is Bob, signing off.
Until next time.
Popular Podcasts
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.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.