Hacker News: ddp26

New comment by ddp26 in "Princeton mandates proctoring for in-person exams, upending 133 year precedent"

ddp26 — Wed, 13 May 2026 21:05:11 +0000

Stanford has this policy too. Students get livid when proctoring is proposed, even though cheating is rampant (afaict)

Run Agents Twice

ddp26 — Wed, 13 May 2026 15:46:05 +0000

Article URL: https://futuresearch.ai/blog/run-agents-twice/

Comments URL: https://news.ycombinator.com/item?id=48123448

Points: 6

# Comments: 0

New comment by ddp26 in "Looking at the data behind prediction markets"

ddp26 — Sat, 09 May 2026 00:51:42 +0000

Author here. Great point, and I think this is due to what another commenter points out, that the questions are different.

The right test of this is to take the _same_ markets that run for 90+ days, and check accuracy 90 days out vs 30 days out. I've done this on other prediction market datasets, though not on Kalshi and Polymarket, and found that forecasts are in fact more accurate 30 days out.

I agree that if they weren't, that would be incredibly suspicious!

New comment by ddp26 in "Looking at the data behind prediction markets"

ddp26 — Sat, 09 May 2026 00:46:36 +0000

Author here. Hal Varian pointed me to this 1992 paper, which I think is still considered the canonical empirical piece on what is actually going on in trading behavior that leads to accuracy (or not): https://www.jstor.org/stable/2117471

New comment by ddp26 in "Looking at the data behind prediction markets"

ddp26 — Sat, 09 May 2026 00:44:51 +0000

Yeah. People have put together a Prediction Market Database [1] (in a Google sheet), I think it's pretty well sourced and shows a good number of both real money and play money prediction markets from before 2002.

DARPA did have a big role though, too.

[1] https://docs.google.com/spreadsheets/d/1vGjnJPxdnBKwag3Q9Uy_...

New comment by ddp26 in "Looking at the data behind prediction markets"

ddp26 — Sat, 09 May 2026 00:42:17 +0000

It's true they are "just" summarizing current knowledge. But there are better and worse summaries of current knowledge!

Some summaries, like on some prediction markets, have objective accuracy that is much better than chance.

New comment by ddp26 in "Looking at the data behind prediction markets"

ddp26 — Sat, 09 May 2026 00:39:44 +0000

Author here. Agree, and I wrote in that section "Absolute accuracy is hard to compare across markets on one platform, and across platforms, because different forecasting questions have different difficulties. I addressed this by tracking similar markets on a single platform over time."

Even doing this, it's not apples-to-apples. One thing is, in this article, I filter only to "interesting" markets, so that controls for the % that are "easy" as you describe.

New comment by ddp26 in "How OpenAI delivers low-latency voice AI at scale"

ddp26 — Mon, 04 May 2026 23:08:33 +0000

Yeah, the question in the title can be answered: "by using gpt-4o, a model 2 years behind the frontier, to serve audio responses"

New comment by ddp26 in "Claude Opus 4.7"

ddp26 — Fri, 17 Apr 2026 02:20:08 +0000

Training window cutoff is Jan 2026, when Opus 4.6 was Aug 2025. That quite a lot of new world knowledge.

I think Anthropic is worth $100B more than last week

ddp26 — Thu, 09 Apr 2026 15:33:59 +0000

Article URL: https://futuresearch.ai/anthropic-30b-arr-ipo-valuation/

Comments URL: https://news.ycombinator.com/item?id=47705082

Points: 9

# Comments: 0

New comment by ddp26 in "Lichess and Take Take Take Sign Cooperation Agreement"

ddp26 — Thu, 09 Apr 2026 14:33:19 +0000

The free open source model does have its competitive advantages!

New comment by ddp26 in "Muse Spark: Scaling towards personal superintelligence"

ddp26 — Wed, 08 Apr 2026 16:26:57 +0000

The second paragraph starts "Muse Spark is the first step on our scaling ladder and the first product of a ground-up overhaul of our AI efforts. To support further scaling, we are making strategic investments..."

This article is about Meta, not about the user. Who signs off on these? Is the intended audience other people at Meta, not the user?

New comment by ddp26 in "A forecast of the fair market value of SpaceX's businesses"

ddp26 — Thu, 02 Apr 2026 18:56:26 +0000

Got a source on this? I didn't take into account in this forecast that public markets could be very inefficient in this way.

New comment by ddp26 in "A forecast of the fair market value of SpaceX's businesses"

ddp26 — Thu, 02 Apr 2026 18:55:07 +0000

There is actually a real bull case for xAI (that I don't endorse), e.g. from people who think that chips & computer is the main determiner of model quality. xAI may plausibly soon have the biggest training apparatus of anyone.

I think talent is more important than compute, as I wrote in my Jan 2026 predictions that Anthropic would end up on top this year: https://futuresearch.ai/blog/forecasting-top-ai-lab-2026/

New comment by ddp26 in "A forecast of the fair market value of SpaceX's businesses"

ddp26 — Thu, 02 Apr 2026 17:54:02 +0000

As I wrote in the piece, I'm extremely skeptical that xAI should be valued as if it is a frontier lab.

But as you say, going back to the xAI + SpaceX merger, analysts consistently seem to value it as if it is, so I predict the public will too, at IPO time.

New comment by ddp26 in "A forecast of the fair market value of SpaceX's businesses"

ddp26 — Thu, 02 Apr 2026 17:35:37 +0000

Yeah, I might have stated this poorly. In the forecast it's just a question of expected value, I don't give almost any probability to "Starship is worthless".

My 50% CI on Starship's fair market value at IPO time is $123b - $227b, with a 80% CI even wider, not based on my own modeling, but based on anchoring to analysts that give credible arguments.

New comment by ddp26 in "A $1.75T IPO Would Be Overpaying 30% for SpaceX"

ddp26 — Thu, 02 Apr 2026 17:27:33 +0000

I read your comment as being glib, but in forecasting this I was really puzzled how much to anchor to how analysts tend to value these businesses.

I ended up largely deferring to them, e.g. predicting the public will value xAI at $258 billion ($222b - $310b) at time of IPO, even though I've elsewhere been skeptical that xAI should be valued like a frontier AI lab.

It's a keynesian beauty contest

New comment by ddp26 in "A forecast of the fair market value of SpaceX's businesses"

ddp26 — Thu, 02 Apr 2026 17:17:28 +0000

Yeah, it's wild. But it's not like the P/E should be 30, what do you think would be fair?

That's the thing about SpaceX, some businesses are real businesses that can be modeled in normal ways, like the government launch contracts, and to some degree starlink.

Others, like ~all of xAI, and the starship stuff, are being valued completely independent of revenue. I predict the IPO investors will generally follow the analysis consensus today with those eye-popping numbers.

A forecast of the fair market value of SpaceX's businesses

ddp26 — Thu, 02 Apr 2026 17:00:11 +0000

Article URL: https://futuresearch.ai/spacex-ipo-valuation/

Comments URL: https://news.ycombinator.com/item?id=47617027

Points: 100

# Comments: 205

New comment by ddp26 in "My minute-by-minute response to the LiteLLM malware attack"

ddp26 — Thu, 26 Mar 2026 21:17:19 +0000

Yeah, but uvx has this thing where it can automatically build the latest environment, and pull the latest (unpinned) version, right?