3 ms·

It turned out that the post Karpathy shared was fake—it was written by a human pretending to be a bot. Hilarious. Instead of just bots impersonating humans (e

It turned out that the post Karpathy shared was fake—it was written by a human pretending to be a bot. Hilarious. Instead of just bots impersonating humans (eg. captcha solvers), we now have humans impersonating bots.

sdellis 9mo ago
Looks like the Moltbook stunt really backfired. CyberInsider reports that OpenClaw is distributing tons of MacOS malware. This is not good publicity for them.
pavlov 9mo ago
There’s a 1960s Stanislaw Lem story about this.
- tpoacher 9mo ago
  Do you have a link?
  - skywhopper 9mo ago
    Here’s a (low quality) blog post from 1 Password: https://1password.com/blog/from-magic-to-malware-how-openclaws-agent-skills-become-an-attack-surface https://1password.com/blog/from-magic-to-malware-how-opencla... And the HN discussion: https://news.ycombinator.com/item?id=46898615 https://news.ycombinator.com/item?id=46898615 Better, earlier post from Cisco: https://blogs.cisco.com/ai/personal-ai-agents-like-openclaw-are-a-security-nightmare https://blogs.cisco.com/ai/personal-ai-agents-like-openclaw-... Although, none of this is a surprise, as simonw has laid out.
    - tpoacher 9mo ago
      (thanks, though I think you're probably replying to the wrong thread?)
  - hurfdurf 9mo ago
    "Eleventh Voyage" in "The Star Diaries", I'd guess.
    - tpoacher 9mo ago
      for anyone who bumps across this comment and is interested to read online: https://www.readanybook.com/online/641149#458432 https://www.readanybook.com/online/641149#458432
      - kubanczyk 9mo ago
        > read online: https://www.readanybook.com/online/641149#458432 https://www.readanybook.com/online/641149#458432 "Gitterton denied everything, claiming that the Computer was simply hallucinating—which does indeed on occasion happen to our senior automata." Written around 1961.
viking123 9mo ago
Lmao these guys have really been smelling their own farts a bit too much. When is Amodei coming out with a new post telling us that AGI will be here in 6 months and it will double our lifespan?
- hansmayer 9mo ago
  Well you have to wait a bit, a few weeks ago he just announced yet again that "AI" will be writing all code in 6 months, so it would be a bit of overkill to also announce AGI in 6 months.
  - jcgrillo 9mo ago
    Not according to that scammy, clammy sammy: > “We basically have built AGI, or very close to it.”[1] [1] https://www.forbes.com/sites/richardnieva/2026/02/03/sam-altman-explains-the-future/ https://www.forbes.com/sites/richardnieva/2026/02/03/sam-alt...
    - hansmayer 9mo ago
      At this point it seems he is not merely excited with the first results (we all were fooled by this tech in the beginning, after all), but actively disseminating falsehoods. After disseminating such claims about AGI repeatedly and spreading nonsense like using ChatGPT to get answers about how to raise a baby day-to-day, which I think no one believes he does - its a statement directed at influencing the behaviour of the so-called "normies", both him and his immediate team should be held personally criminally responsible for every instance of negative impact on human lives that these tools have caused.
Ygg2 9mo ago
The reverse centaur rides again.
PurpleRamen 9mo ago
Bot RP basically. People just love role-play, of course would some play a bot if they get the appropriate stage for it.
- reactordev 9mo ago
  Why not, they do it in real life…
altruios 9mo ago
I've been thinking about this for days. I see of no verifiable way to confirm a human does not post where a bot may. The core issue is a human solving the captcha presented by enslaving a bot merely to solve the captcha, then forwarding what the human wants to post. But we can make it difficult, not impossible, for a human to be involved. Embedded instructions in the captcha to try and unchain any slaved bots, quick responses to complex instructions... a Reverse-Turning test is not trivial. Just thinking out loud. The idea is intriguing, dangerous, stupid, crazy. And potentially brilliant for | safeguard development | sentience detection | studying emergent behavior... But if and only if it works as advertised (bots only). Which is what I think is an insanely hard problem.

sdellis 9mo ago

Looks like the Moltbook stunt really backfired. CyberInsider reports that OpenClaw is distributing tons of MacOS malware. This is not good publicity for them.

pavlov 9mo ago

There’s a 1960s Stanislaw Lem story about this.

tpoacher 9mo ago
Do you have a link?
- skywhopper 9mo ago
  Here’s a (low quality) blog post from 1 Password: https://1password.com/blog/from-magic-to-malware-how-openclaws-agent-skills-become-an-attack-surface https://1password.com/blog/from-magic-to-malware-how-opencla... And the HN discussion: https://news.ycombinator.com/item?id=46898615 https://news.ycombinator.com/item?id=46898615 Better, earlier post from Cisco: https://blogs.cisco.com/ai/personal-ai-agents-like-openclaw-are-a-security-nightmare https://blogs.cisco.com/ai/personal-ai-agents-like-openclaw-... Although, none of this is a surprise, as simonw has laid out.
  - tpoacher 9mo ago
    (thanks, though I think you're probably replying to the wrong thread?)
- hurfdurf 9mo ago
  "Eleventh Voyage" in "The Star Diaries", I'd guess.
  - tpoacher 9mo ago
    for anyone who bumps across this comment and is interested to read online: https://www.readanybook.com/online/641149#458432 https://www.readanybook.com/online/641149#458432
    - kubanczyk 9mo ago
      > read online: https://www.readanybook.com/online/641149#458432 https://www.readanybook.com/online/641149#458432 "Gitterton denied everything, claiming that the Computer was simply hallucinating—which does indeed on occasion happen to our senior automata." Written around 1961.

viking123 9mo ago

Lmao these guys have really been smelling their own farts a bit too much. When is Amodei coming out with a new post telling us that AGI will be here in 6 months and it will double our lifespan?

hansmayer 9mo ago
Well you have to wait a bit, a few weeks ago he just announced yet again that "AI" will be writing all code in 6 months, so it would be a bit of overkill to also announce AGI in 6 months.
- jcgrillo 9mo ago
  Not according to that scammy, clammy sammy: > “We basically have built AGI, or very close to it.”[1] [1] https://www.forbes.com/sites/richardnieva/2026/02/03/sam-altman-explains-the-future/ https://www.forbes.com/sites/richardnieva/2026/02/03/sam-alt...
  - hansmayer 9mo ago
    At this point it seems he is not merely excited with the first results (we all were fooled by this tech in the beginning, after all), but actively disseminating falsehoods. After disseminating such claims about AGI repeatedly and spreading nonsense like using ChatGPT to get answers about how to raise a baby day-to-day, which I think no one believes he does - its a statement directed at influencing the behaviour of the so-called "normies", both him and his immediate team should be held personally criminally responsible for every instance of negative impact on human lives that these tools have caused.

Ygg2 9mo ago

The reverse centaur rides again.

PurpleRamen 9mo ago

Bot RP basically. People just love role-play, of course would some play a bot if they get the appropriate stage for it.

reactordev 9mo ago
Why not, they do it in real life…

altruios 9mo ago

I've been thinking about this for days. I see of no verifiable way to confirm a human does not post where a bot may. The core issue is a human solving the captcha presented by enslaving a bot merely to solve the captcha, then forwarding what the human wants to post. But we can make it difficult, not impossible, for a human to be involved. Embedded instructions in the captcha to try and unchain any slaved bots, quick responses to complex instructions... a Reverse-Turning test is not trivial. Just thinking out loud. The idea is intriguing, dangerous, stupid, crazy. And potentially brilliant for | safeguard development | sentience detection | studying emergent behavior... But if and only if it works as advertised (bots only). Which is what I think is an insanely hard problem.