June 25, 2023 • 8 mins
This is an essay readout using an AI voice clone of NFX general partner, James Currier. The AI revolution just ratcheted up the speed bar for startups. You think you’re moving fast enough. You’re not. Speed is the number one advantage of a startup. But now that generative AI here, your definition of speed has to increase 10x.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:04):
This is James Currier, general partnered in FX,as investors and builders were optimistic about
the future of AI.
One experiment among many we've done is we'vetrained AI avatars of our own voices to read
out the essays we write.
So this is my real voice you're hearing, butThis is now my AI voice.
I'd been created by NFX to provide essayreadouts moving forward.
(00:27):
As you'll hear, I am still a pale imitation ofthe real James Currier, but I'll get better
over time.
And if you have suggestions for me, please letus know.
So to start, this week's article is called Peteand AI.
Let's get started.
You are living in a unique moment, a transitionperiod and you're running out of time.
So much of what we think of as human is goingto be newly located and enabled by software and
(00:49):
silicon.
This is a transition that will sweep acrosssoftware, bio, and every other industry.
Moments like this only come around every 14years or so, and it's a privilege to be here to
see it.
But most founders will squander it and move toGantzla as we've written before.
Speed is the number one advantage of a startup,but now that generative AI is here, your
definition of feet has to increase tenants.
(01:11):
The ideas in this essay are adapted from a talkI've given for years within the NFX skill.
It's one of the most transformative sessions wedo.
We have never published it outside the guildand we're not going to, but here's a small
piece of it.
Because with the pace and opportunity, we seewith generative AI, we feel an obligation to
sound the alarm.
Get out of your own ways.
The hard truth is that no one will be able tomove faster than generative AI itself In the
(01:35):
future, we're going to see smaller teams Morgansoftware and automation controlling the tests
we do every day.
LLMs are exceptionally good at reading andwriting.
Summarizing and ranking, the foundation of mostknowledge work we do today, AI won't be
perfect, but will be faster and more accuratethan us on most days doing most things.
AI won't fear derision from its collegeclassmates.
(01:56):
It will launch the product.
AI won't lay awake at night worrying andtherefore fail to perform due to lack of sleep.
AI won't procrastinate.
AI won't if it's ego bruce and cling to a badidea.
AI won't self sabotage itself because itdoesn't feel worthy.
So humans in the loop need to be fast too,which means they need to be even more clear
thinking and fearless than in the past.
(02:16):
So don't wait.
Do it now.
Don't stand in your own way.
The biggest business risk at this early stageof the AI market cycle is that founders don't
move fast enough into the seams opening in themarket.
The incumbents will have enough time to makethese technologies into features of their
existing businesses.
There are 4 major benefits of the speedstrategy, quality.
Your product gets better faster.
(02:36):
Productivity.
You and your team have more fun accomplishingmore network activation.
Speed and momentum is seductive.
You'll generate more attention from potentialcustomers, employees, press, and investors.
Keep more equity in your company.
Your valuation will go up, and your costs willgo down reducing the need to raise more
capital.
Sounds great.
Right?
(02:56):
So why not more people go faster?
The biggest issue I see is that most people'sspeed bars are just set too long.
You think you're going fast, but you're not,but it's honestly not your fault.
Your speed bar was set too low by the fact thatyou've already been successful.
Most founders we see were good at school.
School is slow based on calendars and loandeadlines, or maybe you were a great employee
(03:17):
at a major corporation.
How fast is a high status corporation move?
You have been trained badly by your environmentto over analyze, to double check, to ensure
everyone has an opinion on every decision.
Maybe you feel like you have a lot to lose.
You don't wanna reset your speed bar because offear you will stray from the processes that
gave you success in the past.
The first step to moving fast is to reset thatbar.
(03:39):
Find a 10 x faster benchmark and reach for it.
Here's an example.
About a decade ago.
We got into the online video gaming businesswithout having built a game before.
Logically, we said, let's hire video gamedevelopers who will know better than us.
Successful video game developers, then told usit would take 12 months to build a really good
game the team could be proud of.
I said, I was thinking of 35 days, not 12months.
(04:02):
They told us it was impossible, but one of myweb engineers, who had never been in the gaming
space, told us he could build a game in 21 daysby himself.
21 days later, he delivered a working gamebuilt in Flash for the win.
That game got 3,000,000 users in 3 weeks.
The gaming people's minds were blown by thePete.
No one builds a game in 21 days, but this webengineer didn't know he couldn't.
(04:24):
He had no context, no industry standards, speedbar holding him back.
Here's the speed bar for most product launchesbefore generative AI.
If you're not experimenting between 20 and 100times per week, you're moving too slow.
If you're not hitting 20 to 100 experiments permonth, you're out of the game.
But with generative AI, open yourself up to theidea that perhaps the speed bar just got even
(04:46):
higher.
Most of you will have to unlearn decades ofslowness.
You'll have to overcome fears that will driveyou to move slower than you should.
We have a few dozen pieces of advice on theFlint that we worked through with our guild
founders, today we're sharing 6.
And it's not the kind of life advice you'reused to.
Most product founders look for speed viaworking more hours.
Were productivity hacks, time managementskills, slack tricks, Beller delegation
(05:10):
principles, apps, and automations, fastertools, But the truth is that AI will figure all
that out and do all those tricks for you and benone of that matters if you don't have the
mindset you need.
Adopt these mindsets to get Pete.
Number 1, what you're building is not arepresentation of you.
How smart you are, how much you'd be loved.
The product is not you.
Just Pete it out the door.
(05:30):
You need to get product in market, see whatworks, what doesn't, see what makes people
uncomfortable, and get them through that cycle.
Watch your competitors closely and borrow thebest ideas.
Don't make it perfect.
Don't spend too much time hunting down specificdata in hopes of building the perfect model if
it comes at the expense of other layers in thegenerative AI stack.
Launch the feature first that the model learnedover time.
(05:52):
Number 2, he were not here to play business oreven to build a product.
The fastest mindset is to think of what you'redoing as running a series of experiments.
This will set you free if you're notexperimenting 20 and 100 times per week you're
moving too slowly.
This is even more pressing given the speed.
Capability and ubiquity of generative AI.
He experimenters who create the 3rd wave of AIcompanies the visionary stuff, the stuff we
(06:15):
can't imagine yet are going to discover whatworks.
AI has made it even more clear.
The experimenters are going to win.
Number 3, to get away from your fear, allowyourself to borrow from others.
You don't need to invent everything.
Pete things at work don't debate it so much.
Just get it up and get the data.
If you need a calendar feature, copy the bestcalendar, Facebook copied friendster, Google
(06:37):
copied everything, It's okay.
Eventually, you will create something uniquebecause you have a unique mind.
Number Fua, the best things that happened in mybusiness career happened after I finally said,
fuck it.
We were running out of money.
Our backs were against the mill.
I gave up and I did the right thing.
I got out of my own way.
I built a product in Morgan wanted, you canstart with this mindset.
(06:58):
Don't wait for the last 30 or 60 days ofcapital to adopt that mindset.
Number 5, particularly outstanding foundershave an unusual drive coming from somewhere
inside their personality for some of them.
That drive comes from feeling unworthy and thusneeding to prove something.
But the paradox of this personality is thatthese people also do things to self sabotage.
All humans do this to some degree, but forfounders in the age of ai, there really isn't
(07:22):
time for it.
If you have this, you need to be aware of itand you have to get past it.
Have your co founders call you out on it.
This is a process.
But there's enough on the internet now toeducate yourself about this state of mind.
Spend some time thinking about the subtle waysthat you might be hampering your own success.
And finally, number 6, if you're not going togo fast for yourself, then here's a mental
(07:42):
hack.
Go fast and win for someone else.
Go fast for your employees.
Your investors, your users, your family, theyare depending on you.
And the fact is you owe it to them.
Don't hurt them with your procrastination orunwillingness to decide and focus.
In conclusion.
There is a lot more to say about going fast instartups, but this is a good beginning to
summarize.
It's not about more hours.
(08:02):
It's about your mindset.
Particularly in the age of AI.
Thanks for listening to this week's essayreadout of speed and AI.
As a reminder.
This is still an AI imitation of James Currierspeaking.
We are having fun experimenting with new tools,but would love to hear your feedback.
Email us at qed@nfx.com.
Subscribe to the NFX podcast on your favoritelistening platform.
(08:25):
Share with you, network, and founder friends,and stay tuned for weekly essays and much more.
This is James Currier, general partnered in FX,as investors and builders were optimistic about
the future of AI.
One experiment among many we've done is we'vetrained AI avatars of our own voices to read
out the essays we write.
So this is my real voice you're hearing, butThis is now my AI voice.
I'd been created by NFX to provide essayreadouts moving forward.
(00:27):
As you'll hear, I am still a pale imitation ofthe real James Currier, but I'll get better
over time.
And if you have suggestions for me, please letus know.
So to start, this week's article is called Peteand AI.
Let's get started.
You are living in a unique moment, a transitionperiod and you're running out of time.
So much of what we think of as human is goingto be newly located and enabled by software and
(00:49):
silicon.
This is a transition that will sweep acrosssoftware, bio, and every other industry.
Moments like this only come around every 14years or so, and it's a privilege to be here to
see it.
But most founders will squander it and move toGantzla as we've written before.
Speed is the number one advantage of a startup,but now that generative AI is here, your
definition of feet has to increase tenants.
(01:11):
The ideas in this essay are adapted from a talkI've given for years within the NFX skill.
It's one of the most transformative sessions wedo.
We have never published it outside the guildand we're not going to, but here's a small
piece of it.
Because with the pace and opportunity, we seewith generative AI, we feel an obligation to
sound the alarm.
Get out of your own ways.
The hard truth is that no one will be able tomove faster than generative AI itself In the
(01:35):
future, we're going to see smaller teams Morgansoftware and automation controlling the tests
we do every day.
LLMs are exceptionally good at reading andwriting.
Summarizing and ranking, the foundation of mostknowledge work we do today, AI won't be
perfect, but will be faster and more accuratethan us on most days doing most things.
AI won't fear derision from its collegeclassmates.
(01:56):
It will launch the product.
AI won't lay awake at night worrying andtherefore fail to perform due to lack of sleep.
AI won't procrastinate.
AI won't if it's ego bruce and cling to a badidea.
AI won't self sabotage itself because itdoesn't feel worthy.
So humans in the loop need to be fast too,which means they need to be even more clear
thinking and fearless than in the past.
(02:16):
So don't wait.
Do it now.
Don't stand in your own way.
The biggest business risk at this early stageof the AI market cycle is that founders don't
move fast enough into the seams opening in themarket.
The incumbents will have enough time to makethese technologies into features of their
existing businesses.
There are 4 major benefits of the speedstrategy, quality.
Your product gets better faster.
(02:36):
Productivity.
You and your team have more fun accomplishingmore network activation.
Speed and momentum is seductive.
You'll generate more attention from potentialcustomers, employees, press, and investors.
Keep more equity in your company.
Your valuation will go up, and your costs willgo down reducing the need to raise more
capital.
Sounds great.
Right?
(02:56):
So why not more people go faster?
The biggest issue I see is that most people'sspeed bars are just set too long.
You think you're going fast, but you're not,but it's honestly not your fault.
Your speed bar was set too low by the fact thatyou've already been successful.
Most founders we see were good at school.
School is slow based on calendars and loandeadlines, or maybe you were a great employee
(03:17):
at a major corporation.
How fast is a high status corporation move?
You have been trained badly by your environmentto over analyze, to double check, to ensure
everyone has an opinion on every decision.
Maybe you feel like you have a lot to lose.
You don't wanna reset your speed bar because offear you will stray from the processes that
gave you success in the past.
The first step to moving fast is to reset thatbar.
(03:39):
Find a 10 x faster benchmark and reach for it.
Here's an example.
About a decade ago.
We got into the online video gaming businesswithout having built a game before.
Logically, we said, let's hire video gamedevelopers who will know better than us.
Successful video game developers, then told usit would take 12 months to build a really good
game the team could be proud of.
I said, I was thinking of 35 days, not 12months.
(04:02):
They told us it was impossible, but one of myweb engineers, who had never been in the gaming
space, told us he could build a game in 21 daysby himself.
21 days later, he delivered a working gamebuilt in Flash for the win.
That game got 3,000,000 users in 3 weeks.
The gaming people's minds were blown by thePete.
No one builds a game in 21 days, but this webengineer didn't know he couldn't.
(04:24):
He had no context, no industry standards, speedbar holding him back.
Here's the speed bar for most product launchesbefore generative AI.
If you're not experimenting between 20 and 100times per week, you're moving too slow.
If you're not hitting 20 to 100 experiments permonth, you're out of the game.
But with generative AI, open yourself up to theidea that perhaps the speed bar just got even
(04:46):
higher.
Most of you will have to unlearn decades ofslowness.
You'll have to overcome fears that will driveyou to move slower than you should.
We have a few dozen pieces of advice on theFlint that we worked through with our guild
founders, today we're sharing 6.
And it's not the kind of life advice you'reused to.
Most product founders look for speed viaworking more hours.
Were productivity hacks, time managementskills, slack tricks, Beller delegation
(05:10):
principles, apps, and automations, fastertools, But the truth is that AI will figure all
that out and do all those tricks for you and benone of that matters if you don't have the
mindset you need.
Adopt these mindsets to get Pete.
Number 1, what you're building is not arepresentation of you.
How smart you are, how much you'd be loved.
The product is not you.
Just Pete it out the door.
(05:30):
You need to get product in market, see whatworks, what doesn't, see what makes people
uncomfortable, and get them through that cycle.
Watch your competitors closely and borrow thebest ideas.
Don't make it perfect.
Don't spend too much time hunting down specificdata in hopes of building the perfect model if
it comes at the expense of other layers in thegenerative AI stack.
Launch the feature first that the model learnedover time.
(05:52):
Number 2, he were not here to play business oreven to build a product.
The fastest mindset is to think of what you'redoing as running a series of experiments.
This will set you free if you're notexperimenting 20 and 100 times per week you're
moving too slowly.
This is even more pressing given the speed.
Capability and ubiquity of generative AI.
He experimenters who create the 3rd wave of AIcompanies the visionary stuff, the stuff we
(06:15):
can't imagine yet are going to discover whatworks.
AI has made it even more clear.
The experimenters are going to win.
Number 3, to get away from your fear, allowyourself to borrow from others.
You don't need to invent everything.
Pete things at work don't debate it so much.
Just get it up and get the data.
If you need a calendar feature, copy the bestcalendar, Facebook copied friendster, Google
(06:37):
copied everything, It's okay.
Eventually, you will create something uniquebecause you have a unique mind.
Number Fua, the best things that happened in mybusiness career happened after I finally said,
fuck it.
We were running out of money.
Our backs were against the mill.
I gave up and I did the right thing.
I got out of my own way.
I built a product in Morgan wanted, you canstart with this mindset.
(06:58):
Don't wait for the last 30 or 60 days ofcapital to adopt that mindset.
Number 5, particularly outstanding foundershave an unusual drive coming from somewhere
inside their personality for some of them.
That drive comes from feeling unworthy and thusneeding to prove something.
But the paradox of this personality is thatthese people also do things to self sabotage.
All humans do this to some degree, but forfounders in the age of ai, there really isn't
(07:22):
time for it.
If you have this, you need to be aware of itand you have to get past it.
Have your co founders call you out on it.
This is a process.
But there's enough on the internet now toeducate yourself about this state of mind.
Spend some time thinking about the subtle waysthat you might be hampering your own success.
And finally, number 6, if you're not going togo fast for yourself, then here's a mental
(07:42):
hack.
Go fast and win for someone else.
Go fast for your employees.
Your investors, your users, your family, theyare depending on you.
And the fact is you owe it to them.
Don't hurt them with your procrastination orunwillingness to decide and focus.
In conclusion.
There is a lot more to say about going fast instartups, but this is a good beginning to
summarize.
It's not about more hours.
(08:02):
It's about your mindset.
Particularly in the age of AI.
Thanks for listening to this week's essayreadout of speed and AI.
As a reminder.
This is still an AI imitation of James Currierspeaking.
We are having fun experimenting with new tools,but would love to hear your feedback.
Email us at qed@nfx.com.
Subscribe to the NFX podcast on your favoritelistening platform.
(08:25):
Share with you, network, and founder friends,and stay tuned for weekly essays and much more.
Popular Podcasts
United States of Kennedy
United States of Kennedy is a podcast about our cultural fascination with the Kennedy dynasty. Every week, hosts Lyra Smith and George Civeris go into one aspect of the Kennedy story.
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.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com