Plus one of the big ways we evaluate the intelligence of other species is trying to see if they have theory of mind, which is intrinsically linked to social intelligence.

Edit: Ah, the person you replied to also invoked special circuitry.

I'd be interested in if these claims fit with places where "traditional" births/care systems are more common, i.e. places where births are primarily supported by the elder women of the community rather than formally educated medical professionals. Though, such places are also less likely to be reached by researchers.

I'm inclined to think there isn't much of an association becauss investors don't seem very concerned with morality, but I know ~dozen developers that either switched to, or started using Claude in the past month or so, while not knowing anyone that uses Codex.

On top of that, sports betting inevitably leads into match fixing, threatening of players etc.

Plus, all of the reports of people being bet on receiving threats.

It's like we've forgotten all about why gambling is so heavily regulated in most of the world. It always metastizes into a cancer upon society unless very carefully and strictly confined away.

Why would they actively subsidize the ticking timebomb? When OpenClaw has an especially large security incident, Anthropic will probably be affected just for the association.

Like, right alongside this post on the front page, we have a post about a relatively serious privilege escalation vulnerability in OpenClaw.

For most satellites/space stations, you need a proper payload deployment mechanism. The pez dispenser mechanism was chosen because opening the entire payload bay and closing it back up for reentry is a tough problem. For now it has been put aside to focus on the goals for Artemis, but that also means not being able to launch stuff other than Starlink.

Starship is currently still stuck in development hell, Musk is already backing off from his Mars plans, SpaceX is moving to distractions and going public (something they previously claimed would not be done).

To me, these moves do not suggest confidence in Starship's ability to live up to its advertised capabilities.

We should be using these capabilities to allow ourselves to work on harder problems. In science, there are a lot of tasks that require a low, but non-zero amount of intelligence and aren't really the most interesting part of science. Many of these tasks limit how much work can actually be done. Automate them, and you can dramatically increase your capabilities and focus on the actual science work.

It's a pretty strange issue, makes me feel like the 1M context model was actually a downgrade, but it's probably something weird about the state of its memory document. I wasn't even very deep into the context.

We've managed to optimize execution of the simulation enough that brute-force search is a viable option, but giving an agent some background on how we tune those parameters on intuition and some physical reasoning, and a means to run tests and retrieve resulting statstics, works surprisingly well.

I see it as essentially a hyperparameter search that is more capable of finding and exploiting implicit constraints in a system.