andromaton

4 ms·

1.
King Ludd(astralcodexten.com)
3 points by andromaton 6d ago|0 comments
2.
Flawed benchmarks (epoch.ai benchmark registry filtered by flawed)(epoch.ai)
2 points by andromaton 8d ago|0 comments
3.
WeirdML v3(htihle.github.io)
2 points by andromaton 8d ago|0 comments
4.
by andromaton 27d ago
Split personality A) pierce reality B) stocks
5.
The Foothills Of Bay Area House Party(astralcodexten.com)
4 points by andromaton 28d ago|3 comments
6.
by andromaton 2mo ago
AI2027.com puts this around early 2027 and treats the loss of legible chain-of-thought as a major negative turning point for alignment.
7.
by andromaton 2mo ago
to those that want to "punish" the uk ai security institute: these failures are good for all of society.
8.
by andromaton 2mo ago
Sorry for his loss, and for everybody helped by her in any way. However ineffective a comment like this is
9.
by andromaton 2mo ago
Pixel accurate is easy to say, hard to do, over-specific, and counterproductive.
10.
by andromaton 2mo ago
We've known artificial intelligence will do unexpected things since the 90s and that it can do difficult things since 2025. Pausing worldwide not easy but we do harder things all the time.
11.
by andromaton 2mo ago
impressed at the look (not the geometry) of 3.6 lol at flash-lite could not find your posts about them
12.
But what is quantum computing? (3blue1brown 2025)(youtube.com)
9 points by andromaton 3mo ago|2 comments
13.
by andromaton 3mo ago
Extremely useful
14.
by andromaton 3mo ago
A lower key Putnam moment
15.
Games Between Programs: The Ruliology of Competition(writings.stephenwolfram.com)
36 points by andromaton 4mo ago|4 comments
16.
by andromaton 4mo ago
Self-replication requires a minimum amount of information plus machinery. Consciousness may have a similar threshold: below some level of self-modeling, nothing; above it, degrees.
17.
Latency Numbers Every Programmer Should Know (2026)(gist.github.com)
1 points by andromaton 4mo ago|1 comments
18.
by andromaton 4mo ago
Now includes multi-second llm latencies
19.
by andromaton 4mo ago
Since yesterday.
20.
by andromaton 4mo ago
Claude FM: heartfelt Gemini: psycho Grok: incompetent OpenAI: by the book
21.
by andromaton 5mo ago
If priced like other Anthropic models, Mythos will make vulnerability discovery a lot more accessible. The author compares it to AISLE, ZeroPath, and OpenAI’s Codex Security. AISLE and ZeroPath are much more expensive. OpenAI’s Codex Securi
22.
by andromaton 5mo ago
Dave Barry also insists he is not dead. https://davebarry.substack.com/p/death-by-ai It seems it's quite exhausting.
23.
by andromaton 5mo ago
Also too many lands and hits.
24.
by andromaton 5mo ago
Related https://news.ycombinator.com/item?id=47838508#47839757
25.
by andromaton 5mo ago
The ux of copilot driving Claude beats Claude Code handily. I never understood the low visibility. Expensive ram is annoying. I don't look forward to expensive ai.
26.
by andromaton 5mo ago
- New sign-ups paused for Copilot Pro, Pro+, and Student plans. - Individual plan limits tightened. - Opus models removed from Pro plan.
27.
Announcement: Changes to GitHub Copilot Individual Plans(github.com)
10 points by andromaton 5mo ago|3 comments
28.
by andromaton 6mo ago
A comment from someone who knows or knew the author or was part of the project sharing details that makes readers feel like they've just been handed backstage passes.
29.
by andromaton 6mo ago
A highly voted comment that seems insightful if you don't know the domain but amateurish if you do.
30.
by andromaton 6mo ago
A comment noting the article is from YYYY and asking if the mods can add (YYYY) to the title.

More ›