It’s meant to be an attempt at humor/sarcasm/engagement bait I think.

I’ve done this (asked a frontier model red team my server), and it just gave me false alarms that I went investigating and weren’t really issues (would’ve been an issue if I had a very specific configuration, but not generally). They produce a lot of noise, so you still have to somewhat understand what you’re looking at/for. It didn’t find any legitimate issues, which is good I guess.

Governments are captured by these companies. Any regulation would undoubtedly benefit these companies at the peoples expense. Right now, the best solution is probably open research and open weight models to keep the field level.

If it’s like Tesla stock, it will behave completely irrationally. Though, I guess most of the stock market is irrational now.

My favorite way is Borscht. They’re pretty good grilled, then cut up, and put in salad too.

Windows and the fascist company behind it needs to die.

Some big tech companies pay that, theoretically, in total compensation for entry level. These companies make about $1 million per employee.

I think it’s the same in all developed nations; constantly needing more skills to achieve the same standard of living. I think a lot of it is from nearly all resources getting more expensive to extract (oil, wood, iron, etc) due to us having already extracted all the low-hanging-fruit, and needing to move on to more resource-intensive methods like offshore-drilling, fracking, importing lumber long distances from harsher climates. The other drivers are the attacks on labor and executives/shareholders taking more profits for themselves instead of paying their workers more.

I think I read the RLHF kind of makes these logprobs completely unusable too.

I think it’s driven by the investors. In the case of big tech, the large institutional investors are rewarding companies any time they say “AI” and lay off workers. In the case of startups, VCs are almost exclusively investing in startups that use “AI,” and have a lean or offshore workforce.

I find it hard to believe the true numbers are this low. Every job posting gets many hundreds or even thousands of applicants. It’s a shame so much talent is wasted by so many people being unemployed and doing “unproductive” things like spending months applying to jobs.

IDK about most. But, I’ve seen many OS contributors say they’re looking for work. Seen one recently saying he won’t be contributing much to the project anymore because he’s housing-insecure. Seen maintainers for popular projects get laid off and are now looking for work. Seen people with 10+ and 20+ years of experience not being able to find a job after many months.

Does Snowflake still work in China? Thought I read they’re now able to detect and block it.

There’s a lot of indication that LLMs are peaking. It’s taking exponentially more compute and data to get incremental improvements. A lot of people are saying OpenAI’s new model is a regression (I don’t know, I haven’t really played with the new model much). More foundational breakthroughs need to be made, and these kinds of breakthroughs are often the result of “eureka” moments which can’t be manifested by just throwing more money at the problem. It’s possible it will take decades before someone discovers a major breakthrough (or it could be tomorrow).

China leads the world in scientific publication, even when only taking into account reputable journals and high-impact publications. There’s no doubt in my mind the US will decline further with the current attacks on science and education, and anti-intellectualism in general.