A look inside Mindstate Design Labs’ effort to design drugs that reliably produce specific states of consciousness.

A research-backed AI scenario forecast.

I’ve argued elsewhere that it’s plausible AGI arrives before 2030. That’s a big claim.

As a non-expert, it would be great if there were experts who could tell us what to think.

Unfortunately, there aren’t.

At least two mass extinction events in Earth’s history were likely caused by the devastating effects of nearby supernova explosions. That’s according to a new study by researchers at Keele University in England. The researchers said these super-powerful blasts – caused by the death of a massive star – might have previously stripped our planet’s atmosphere of its ozone, sparked acid rain and exposed life to harmful ultraviolet radiation from the sun. They believe a supernova explosion close to Earth could be to blame for both the late Devonian and Ordovician extinction events, which occurred 372 and 445 million years ago respectively.

Silicon Valley elites are quietly appointing themselves ancestors of the future.

As American fertility rates continue to hit record lows — well below the “replacement rate” from one generation to the next — the pronatalism movement has found a foothold in the tech world.

Elon Musk — who apparently now has 13 children — has openly endorsed the movement, which believes in pumping out babies to fight societal collapse. OpenAI CEO Sam Altman has said, “Of course I’m going to have a big family” and is investing in experimental fertility technology.

Imagine being able to speed up evolution – hypothetically – to learn which genes might have a harmful or beneficial effect on human health. Imagine, further, being able to rapidly generate new genetic sequences that could help cure disease or solve environmental challenges.

Now, scientists have developed a generative AI tool that can predict the form and function of proteins coded in the DNA of all domains of life, identify molecules that could be useful for bioengineering and medicine, and allow labs to run dozens of other standard experiments with a virtual query – in minutes or hours instead of years (or millennia).

Trained on a dataset that includes all known living species – and a few extinct ones – Evo 2 can predict the form and function of proteins in the DNA of all domains of life.

Science Fiction…for now.

I’m not a natural “doomsayer.” But unfortunately, part of my job as an AI safety researcher is to think about the more troubling scenarios.

I’m like a mechanic scrambling last-minute checks before Apollo 13 takes off. If you ask for my take on the situation, I won’t comment on the quality of the in-flight entertainment, or describe how beautiful the stars will appear from space.

I will tell you what could go wrong. That is what I intend to do in this story.

How much progress have physicists made on the 10 millennium problems?

So far their success rate is 1 out of 10.

In the year 2000, physicists created a list of the ten most important unsolved problems in their field. 25 years later, here’s where we are.

This is a low-effort post. I mostly want to get other people’s takes and express concern about the lack of detailed and publicly available plans so far. This post reflects my personal opinion and not necessarily that of other members of Apollo Research. I’d like to thank Ryan Greenblatt, Bronson Schoen, Josh Clymer, Buck Shlegeris, Dan Braun, Mikita Balesni, Jérémy Scheurer, and Cody Rushing for comments and discussion.

I think short timelines, e.g. AIs that can replace a top researcher at an AGI lab without losses in capabilities by 2027, are plausible. Some people have posted ideas on what a reasonable plan to reduce AI risk for such timelines might look like (e.g. Sam Bowman’s checklist, or Holden Karnofsky’s list in his 2022 nearcast), but I find them insufficient for the magnitude of the stakes (to be clear, I don’t think these example lists were intended to be an extensive plan).