forecasting the future at preseen. prev @princeton
San Francisco
- Tweets 464
- Following 478
- Followers 1,045
- Likes 3,942
Veniamin Veselovsky retweeted
.@AskPreseen is so cool and powerful for helping forecast the success of a startup (or of anything) and we are proud to include that functionality in the launch of Idealab.ai today..
Super excited to be partnering with @Bill_Gross and Idealab.ai to integrate Preseen into forecasting start up idea success!
1
6
2,031
Super excited to be partnering with @Bill_Gross and Idealab.ai to integrate Preseen into forecasting start up idea success!
Preseen is now available on @Idealab.ai to put forecasts in the hands of founders deciding what business to start.
Idealab.ai makes more than 1.2 million business ideas freely searchable, drawing on @Bill_Gross's 50-year entrepreneurial experience.
1
9
2,141
Veniamin Veselovsky retweeted
We didn't want to lose to the bot, but we didn't end up beating it by much.
Did human forecasters improve Preseen’s performance?
In the Summer 2026 @metaculus Cup, Scott Eastman Eastman and @rdeneufville worked with the Preseen bot as Symbiosis.
1
4
249
Can two superforecasters + machine outperform machine in forecasting?
When the machine first beat a human in chess there was a point in time when man + machine > machine.
At Preseen we’ve built out one of the best AI systems for forecasting. We wanted to see if two superforecasters working with our system could outperform the system itself.
They ended up narrowly beating the system, but what we learned was fairly surprising.
A few takeaways:
- Being a domain expert doesn’t map onto using an AI system well. Guiding superhuman (or at least on-par) AI requires a different skillset. Effectively working with an AI system requires an additional skill: knowing when and how your own expertise can improve its output.
- Evaluating a centaur approach is non-stationary since the human’s behavior with the agent changes as they learn more about the constraints of the agents. If an agent has a few dimensions a human can interact with, the human’s use of those dimensions can change dramatically.
- Human agreement could reinforce a mistake. On the three questions where overrides cost the most, the humans moved toward the community prediction. What looked like a sensible correction supported by other forecasters ended up hurting performance.
6
5
28
4,306
Veniamin Veselovsky retweeted
Introducing Kardashev-0.7, the world’s first trained swarm made of 32 distinct models
Trained with RL for Population Scaling (RLPS), 32 models organically develop specialization & complementary capabilities, delivering frontier performance at:
- 0.007x ~ 0.02x of the inference cost
- 0.03x of the required memory
Civilization advances through different minds specializing and working together. We’re bringing that principle into AI
@BanburyRoadAI, we’re scaling intelligence by model count, toward civilizations of models that learn to build on one another
450
417
4,811
425,452
I’ve been a fan of James’ work since undergrad! Super excited for him to be joining Preseen as our Chief Scientist.
James Evans (@profjamesevans) has joined Preseen as Chief Scientist.
He's a @UChicago professor and founding director of Knowledge Lab and the Chicago Center for Computational Social Science (C3S2), studying how people and AI generate knowledge and anticipate discoveries.
2
16
1,958
A lot of experts quoting probabilities for p(doom) have little to no forecasting background.
We've repeatedly seen that experts have access to information that's relevant for a forecast, but produce uncalibrated forecasts themselves.
For reference, here is Preseen's forecast for "Will AI cause human extinction by December 31, 2100?": preseen.com/q/7Q94Kyncj4XNzv…
“One of the main lessons to come out of Tetlock’s work is that the best generalist forecasters are usually better at telling the future than domain-specific experts, even when it comes to those experts’ areas of study. This is because forecasting itself is its own science that requires a unique way of thinking, and a person having a great deal of knowledge about a field is no guarantee that they have good epistemological habits. If you’re going to ask, for example, whether American democracy is falling apart, an expert might start by making a superficial comparison between the modern US and the Weimar Republic. Someone with a forecasting mindset would be more likely to begin with a question like “how often do advanced democracies slide into dictatorships?” The answer is not very often.”
Important reminder from @RichardHanania that domain experts are not the best at predicting what happens, even in their own domains.
open.substack.com/pub/richar…
15
1,242
Forecasting science and technology is hard. When Bell Labs unveiled the transistor in 1948, The New York Times gave it a throwaway item on page 46.
This pattern holds across fields with the best-funded programs and most-cited papers seldom being the ones that matter most, and the advances that do are often ones no one saw coming.
I'm excited Preseen will be partnering with the Knowledge Lab at UChicago to incorporate their data and models into Preseen's core forecasting technology to improve our ability to forecast science and technology.
More on this here: blog.preseen.com/p/predictin…
2
3
32
2,276
Veniamin Veselovsky retweeted
We’re partnering with @UChicago's Knowledge Lab to advance AI forecasting for science and technology.
The partnership builds on KLab director @profjamesevans's involvement in Preseen’s founding.
2
3
8
590
Veniamin Veselovsky retweeted
Forecast Peek 👀
How much will Ticketmaster and Live Nation ultimately pay in the FTC and seven states’ ticket-pricing lawsuit?
Preseen’s median forecast is $220M, with a 68.9% chance of more than $100M and 14% above $1B.
1
1
5
180
Veniamin Veselovsky retweeted
I am so excited to share what we have been up to!
TLDR; We built Atlas — the best camera conditioned frame autoregressive transformer in the world.
Finally, I can tick training a frontier model off my bucket list :):)
Tysm & shoutout to my incredible team at @theworldlabs
Introducing Atlas:
The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D.
Model the world, move the camera, and simulate space & time.
14
14
175
13,029
Veniamin Veselovsky retweeted
Beautiful paper on RSI!
RSI will not just be internally for the frontier labs, but will come to the enterprise very soon where frontier models will train specialized task-specific models.
A lot of infrastructure will be built just to enable this for the enterprise. Particularly -
1. Auto-research harnesses
2. Aggregation of research so auto-research agents can learn from past experiments
3. Cheaper training
15
26
240
15,469
Veniamin Veselovsky retweeted
A recent episode of Hard Fork (@nytimes) takes on the much-debated subject of AI forecasting: "I think it’s actually quite interesting and profound that we now have, in some cases, AI systems that are better than humans at making predictions about the future." —@kevinroose
This is something we've been tracking closely in FutureEval, our ongoing benchmark measuring AI forecasting accuracy over time (link in replies).
Thank you to Kevin & @CaseyNewton for covering this topic and to @VminVsky for the shoutout!
2
2
15
978
spc is one of the most talent dense places in sf.
Today we’re announcing SPC Fund IV: $575M to help founders find their life’s work.
A decade of guiding founders through -1 to 0 brings @SPC to $2B in AUM and, more importantly, a 1200 member community of the world’s most talented technologists. We strongly believe that:
→ The person comes before the idea. At -1, generational founders are often busy reading, prototyping, discarding, and earning the right to their own conviction.
→ Ambition is social. The people you surround yourself with in the early days play a big part in inflecting your ambition.
→ Patience compounds. It is better to work on something you have deep conviction in rather than simply sprint at the first idea.
The quiet months we protect at the beginning have produced three funds in the top 10% of their vintages, and companies like @baseten, @GammaApp, @render, @GoodfireAI, @LumaLabsAI, and @profound. We are proud to have worked with all these founders in their -1 phase and beyond.
While we defined -1 to 0, we're no longer bound by it. Fund IV allows us to partner with companies well beyond launch.
We're doing this now because we've never seen a wider gap between what's possible and what's being attempted. Intelligence is abundant. The cost of trying ideas keeps falling, and problems that seemed insurmountable a few years ago are quickly becoming solvable.
To our members, to the founders who trusted us before anyone else did, and to the LPs who had conviction in our own -1 journey: thank you. This milestone belongs to you as much as to us.
To the builders looking to find your life’s work: let’s get started.
2
1
25
5,795
Veniamin Veselovsky retweeted
Every decision is a forecast. Better forecasts mean better decisions.
@VminVsky and Theo are building the best forecasting agents at @AskPreseen.
I spent my career building systems, so I want to be precise about why this is a hard technical problem and not a wrapper.
A language model out of the box is a poor forecaster. It reasons from priors frozen at training, inherits the biases of whatever it retrieves, and will happily give you a confident number with nothing underneath it.
Getting from there to a calibrated probability means solving several problems at once. You need fresh, structured data about the state of the world, which is an ingestion and normalization problem across thousands of heterogeneous sources, not a search query.
You need to force independence between lines of reasoning so errors stay uncorrelated, which Preseen does by running separate agents against primary records and reconciling their estimates the way you would aggregate a panel of analysts.
And you need calibration, which is its own discipline: a system that says 80% has to be right about eight times in ten, measured across hundreds of resolved questions.
Forecasting is one of the rare AI problems with delayed ground truth. Every question eventually resolves, so every forecast becomes a labeled training example.
Preseen has been accumulating that feedback since last fall, on top of a data layer Venia spent months building: crawlers and normalization pipelines across country APIs, satellite imagery, procurement records, and stranger signals like diplomatic flight patterns and international school enrollments.
Last December that machinery flagged unusual Russian flight activity before it showed up anywhere humans could see it. A general-purpose model gets none of this for free. The loop has to be built, and it gets stronger with every resolved question.
The Preseen bot won the FutureEval tournament last fall, first out of 164. It came third out of 1,283 in the Metaculus Cup this spring, ahead of some of the best human forecasters alive. Then in Q2 it won Market Pulse outright, the first time a bot has ever beaten a field of humans to win a Metaculus tournament, and it won by a wide margin. Theo turned $35 into roughly $1.9 million on Kalshi over seven months.
The trend line is very clear that Preseen will soon beat even the best human forecasters. Any decision that matters will need AI forecasting at the table. Everyone wants to know the future.
I am very proud to lead Preseen’s seed round at @spc. They’re hiring.
Over the past year I've been building out Preseen. We’re focused on building the best AI forecasting agent for decision makers.
Every decision relies on a forecast. Too often decisions are poorly made because the underlying forecast is bad. Humans are constrained. But if you can have a system that can forecast anything as well as the best humans, then human decision making will be improved.
AIs are reaching the point where they are better than some of the best human forecasters. This is going to change decision making.
Preseen will help drive that change.
6
6
38
6,726
Veniamin Veselovsky retweeted
AI superforecasters are going to be much better than human ones. Excited about what @VminVsky is building with Preseen.
Over the past year I've been building out Preseen. We’re focused on building the best AI forecasting agent for decision makers.
Every decision relies on a forecast. Too often decisions are poorly made because the underlying forecast is bad. Humans are constrained. But if you can have a system that can forecast anything as well as the best humans, then human decision making will be improved.
AIs are reaching the point where they are better than some of the best human forecasters. This is going to change decision making.
Preseen will help drive that change.
8
3
61
8,988
I’m also thrilled to announce that @spc led our seed putting in $5.6M. The whole team there has been amazing, but I’d like to give a special shoutout to @adityaag and @JPBrebner.
1
18
901