20VC Newsletter - 7th September 2025
Here are the transcripts and top takeaways from 20VC episodes released this last week.
Monday’s episode with Nick Frosst, Co-Founder @ Cohere:
Download the full transcript:
My 5 takeaways:
How Sam Altman Did a Disservice to the Progression of AI
He has made several predictions that were obviously wrong.
He talked about how close AGI is & did a world tour saying AI is going to be an existential threat in 2 years.
It is academically disingenuous & a disservice to the technology he loves.
Not Being an American Tech Company Is an Asset Today
Companies around the world are interested in talking to us because we’re a Canadian company.
The US has shown that they are willing to turn off access to tech based on political reasons.
Not being an American tech company can be an asset today.
Should Countries Fund Their Own LLMs and Why They Are Like Power Plants
It’s a good idea for countries to have infrastructure like power plants within their own country.
Language models are not that dissimilar from power plants.
Should We Have Sovereignty of Models?
Having a model that speaks your country’s language is like building infrastructure for your people.
If you want to empower the people of your country you need a model that understands the cultural context of that country.
Simply using models from the US or China will not achieve that.
Should Every Business Have Forward-Deployed Engineers
It is a great idea.
If you’re selling tech it makes sense to have engineers who set it up & work with clients.
I don’t know if that applies to every business.
Thursday’s episode with Rory O'Driscoll, GP @ Scale, Jason Lemkin, Founder @ SaaStr, and Cliff Obrecht, Founder & COO @ Canva:
Download the full transcript:
My 4 takeaways:
Why the Anthropic Round at $183BN Post Is Not Crazy
If revenue in Jan 2025 ≈ $1BN and they end the year around $9BN. The GAAP revenue will be ≈ $5BN
Assuming trajectory slows down next year (10x to 3x) they will have around $30BN in revenue EOY. The GAAP revenue will be ≈ $20BN.
If you believe the growth will last, you are only buying at 8-9x next year’s revenue.
Are the Wheels Coming off the Scale Acquisition?
There are 3 questionable points:
Talent fallout: People have egos, and people are leaving Meta’s new team.
Asset fit: Meta still buys the best data available & not exclusively from Scale. While Scale’s neutrality has been damaged with Meta’s deal.
Deal structure: $14BN distributed cash. Scale is hollowed & Meta could be forced to write down soon.
If You Are Not Accelerating With AI You Have Failed as a Founder
If you have a billion dollar install base, you have a distribution channel.
If you haven’t accelerated by the end of 2025 you have failed as a founder.
You had 18 months. You have no excuse.
When You Pass On a Company and They Do What They Said They Would, Always Do the Next Round
You do not need any more information: You know you can lean into their execution.
I passed on Omniture in 2003 & bought at 2x the price a year later.
I passed on Box in 2010 & bought it 9 months later.
Friday’s episode with Alex Schultz, CMO & VP Analytics @ Meta:
Download the full transcript:
Let us know what your big takeaways from this week’s shows were in the comments below!
Thank you for reading, and don’t miss the great guests we have next week:
Monday episode: Mati Staniszewski, Co-Founder @ ElevenLabs
Thursday episode: Jason Lemkin, Rory O'Driscoll & Jeff Lawson @ Inertia
Friday episode: Amit Bendov, Co-Founder & CEO @ Gong.io
Thank you for reading 20VC.
This post is public so feel free to share it.
Thanks for reading 20VC! Subscribe for free to receive new posts and support my work.