program_whiz

7 ms·

1.
by program_whiz 4d ago
Surprisingly this is actually rather fitting in terms of time scale. When you consider it took ~10,000 agents 88 hours, or 880,000 hours to solve. That's 14.5 years in agent time of continuous 365/24/7 processing. Of cours
2.
For safety: Keep AI lonely(nonlineartransform.substack.com)
3 points by program_whiz 5d ago|0 comments
3.
AI rent can save humanity(nonlineartransform.substack.com)
2 points by program_whiz 6d ago|1 comments
4.
by program_whiz 6d ago
A modest proposal to save humanity from the alignment problem with AI rent.
5.
by program_whiz 6d ago
Good point, you don't have to -- but my argument is just that removing CoT doesn't make things less safe. Anything CoT can tell you is just a point sample of a probability surface. Having the whole probability surface can alread
6.
by program_whiz 7d ago
Actually, removing CoT might make models safer, because we can analyze the entire landscape of their potential outputs, rather than a point-sample (we'll never know how close we were to "kill all humans"). By inspecting inte
7.
by program_whiz 7d ago
yes to a degree. When I was 16, my idea about "what cutting edge thing to work on" was pointless, I couldn't make a meaningful contribution. Most developer's work is mundane and wrote. Things they think are "cutt
8.
by program_whiz 7d ago
I wrote this in response to people that have been downplaying the threat, or not understand how it would work without terminator-style robots in the streets. I also see people saying "we have a year or two to prepare". That may
9.
We have world ending AI, today(nonlineartransform.substack.com)
2 points by program_whiz 7d ago|1 comments
10.
by program_whiz 7d ago
I wrote about this, as I thought it warranted a deeper look: https://nonlineartransform.substack.com/p/world-ending-ai-to...
11.
by program_whiz 7d ago
What? Shutting down infrastructure, destroying critical records (economic collapse), spoofing / impersonating world leaders (confusion, panic, world war, nuclear event), taking over comms generally (you get a message to evacuate your h
12.
by program_whiz 7d ago
So when AI is also better at coming up with the tasks and already solving the frontier problems? 1M AI agents running full time thinking at 10M times faster than humans, with 100 times the intelligence in each agent? Any human ideas regar
13.
by program_whiz 8d ago
Timing is everything https://nonlineartransform.substack.com/p/ai-swarms-timing-i... Article arguing math is the next "human calculator".
14.
Timing Is Everything: The Terrence Tao Effect(nonlineartransform.substack.com)
1 points by program_whiz 9d ago|0 comments
15.
Agents need capitalism, people need socialism(nonlineartransform.substack.com)
3 points by program_whiz 10d ago|0 comments
16.
by program_whiz 10d ago
A response to this, which I think is reasonable. Its a bit of a fuzzy line, but the likelihood there is "magical maniacal planning" happening here is unlikely (about as unlikely as that planning happening in hidden vector states
17.
Relax about Neural Space(nonlineartransform.substack.com)
1 points by program_whiz 10d ago|2 comments
18.
by program_whiz 10d ago
Comparison of "neural space" and "chain of thought", and why CoT was actually worse.
19.
HF hack alignment: make agents more selfish(nonlineartransform.substack.com)
1 points by program_whiz 11d ago|0 comments
20.
Agents need to be more selfish(nonlineartransform.substack.com)
1 points by program_whiz 11d ago|0 comments
21.
by program_whiz 12d ago
Sorry I disagree with that. This is more like gain of function research. You are trying to develop an agent with the ability to do hacking and the like without having proper safeguards. When it breaks free and causes massive damage, the
22.
by program_whiz 12d ago
equally valid analog: had they merely written a script to do the hugging face exploit, they would go to prison. However, since an "agent" wrote the script for them, nothing happens?
23.
by program_whiz 12d ago
the person who controlled / started it? If open AI had hired a team of 50 hackers to break into hugging face, they would be prosecuted (as would the hackers). If they had written a bot to break into hugging face, the devs and manager
24.
by program_whiz 12d ago
are you arguing AI manufacturers and developers are not aware of possible risks?
25.
by program_whiz 12d ago
The solution is simple: hold anyone who deploys an agent responsible for its behavior. If it commits 10 counts of felony hacking, ouch. If it kills 10 pedestrians by running a red light, ouch. If this is "human level intelligence&qu
26.
by program_whiz 19d ago
the link works for me
27.
by program_whiz 26d ago
Its interesting because I didn't think it was, but then reading the NVIDIA approach, this kind of loop plus generating a program to explain things. Maybe that is AGI? I don't know, but it seems like an additional layer that mayb
28.
by program_whiz 26d ago
Is this AGI? I don't think I can score 100% on ARC AGI.
29.
by program_whiz 1mo ago
AbhiramaVS I think your choice to use exceptions is fine, it matches closely with Java / Python your inspirations. The "show me every error" crowd loves to harp on this issue, but the truth is, its a tradeoff like all others
30.
by program_whiz 1mo ago
I don't expect it to be free of corruption, but I can also acknowledge the systemic forces that are beyond my control which enable the corrupt excesses we see. Unfortunately, until the incentives reverse, it will continue. One of the

More ›