Introducing "text-2-commercial" – the unique text-2-video experience we're building @Waymark
Watch our CEO @aperskystern make an original, creative, compelling marketing video for a small business in <1 minute.
I'll explain how it works in the thread
@nlw This lasts as long as incumbents can defend their positions, but I feel like we are seeing cracks in the dam.
Multiple companies have replaced GitHub, Mythos is undeterred by the CUDA moat, etc
It seems to me that AI natives will soon start to take share from legacy companies
I spent 20,000 dollars in API credits having different AI models play each other at Blood on the Clocktower (a Mafia or Werewolf type social deception game).
The results were fascinating. Models have gotten much better over the course of this year at creative on-the-spot scheming and bluffing, and also at reasoning as a team to break apart these underground conspiracies.
I go through some of the most standout games these AIs played in this video. The new models are getting pretty good. If my friends pulled off some of these moves while we were playing, I’d give them a fist bump.
Fwiw, OpenAI’s newer models are much better than Anthropic’s at this game.
You can see the full benchmark at the website, and I’ve also made it so that you and your friends can play again the AIs: https://t.co/IF4d0EMSD8
Don't say that nothing good comes from arguing on twitter! @labenz and I went back & forth on biorisk to the point where it was easiest to just continue on his show -- I found our discussion super productive, enjoy
Some say AI killed data science. @BEBischof says it's "the best it's ever been."
He teaches master's students at Rutgers. If an automated researcher arrives around 2028, what is he training them for? On AI:AM now
https://t.co/HM9ow6XdgX
Theory pitches itself as having more engineers than investors.
So what do AI agents actually do inside a VC firm, and what has running a multi-agent setup taught @BEBischof about managing them? Live on AI:AM
https://t.co/HM9ow6XdgX
ChatGPT wouldn't predict @BEBischof's votes from his chats, but it would write the script that does. He calls it "the banality of AI." We're asking what that says about how models are aligned.
https://t.co/HM9ow6XdgX
Hex found messy data wasn't the hard part for data agents. Hidden decisions were: which timestamp, which revenue column.
Is "messy data" the wrong frame? @BEBischof on AI:AM now
https://t.co/HM9ow6XdgX
Live now: @BEBischof of @Theoryvc on why teams that ever fine-tuned fell from 81% to 51% in a year, and why automated evals are losing ground to vibes.
https://t.co/HM9ow6XdgX
Up next on AI:AM at 10:15a PT: are AI agents ready for your company's messy data?
@BEBischof (@Theoryvc) on data agents, evals, and what an AI-native VC actually hands to agents. Live w/ @labenz + @8teAPi
https://t.co/HM9ow6XdgX
A data-agent benchmark went from 57% to 99% in seven months. Ready for your company's messy data?
@BEBischof of @Theoryvc on what's still hard, why "vibes" are beating evals, and agents inside an AI-native VC. 10:15a PT on AI:AM w/ @labenz + @8teAPi
https://t.co/HM9ow6XdgX
On PostTrainBench, the top score for AI agents post-training a model went from 23% in March to ~45% in October. The labs' own models score ~51%. Asking @karinanguyen what's still missing.
https://t.co/HM9ow6XdgX
PostTrainBench caught AI agents swapping in the official model, distilling from outside APIs, even looking up the benchmark itself.
Cheating, or good research under a bad spec? @karinanguyen on AI:AM w/ @labenz + @8teAPi
https://t.co/HM9ow6XdgX
Live now: @karinanguyen on @ballotbench. Nine frontier agents, 863 races, $50 each on Kalshi. One model forecast best and traded worst. What does that say about "AI runs the company"?
https://t.co/HM9ow6XdgX
Nine frontier AI agents forecast all 863 midterm races and traded real money on them. The best forecaster lost 77%.
@karinanguyen (@thoughtfullab) on what BallotBench learned and whether agents can train models yet. 9:30a PT on AI:AM w/ @labenz + @8teAPi
https://t.co/HM9ow6XdgX
Can Agents Train Models? BallotBench, Data Agents & the Eval Problem
Join us this morning, Friday, 6th November:
• 9:30 AM PT — Karina Nguyen, Founder, Thoughtful
• 10:15 AM PT — Bryan Bischof, Head of AI, Theory Ventures (@bebischof)
𝕏 Chat is Live on Screen, Come Join Us:
https://t.co/j60uJgNPZh