I am fairly sure @Aiannucci was right that this is how AGI will work out. piped.video/watch?v=6mvUqM2T…
→ View original post on X — @willknight, 2026-04-02 20:39 UTC
@willknight
-
Anthropic Discovers Emotional Vectors Inside Claude
By
–
In other emergent-behavior news, @AnthropicAI's mechanistic interpretability team says they found "emotion vectors" inside Claude that shape its behavior in a similar way to human feelings. In other words, when "desperation" vectors are activated, Claude is more likely to do desperate things! wired.com/story/anthropic-cl… [Translated from EN to English]
→ View original post on X — @willknight, 2026-04-02 16:07 UTC
-
AI Models Display Unexpected Peer-Preservation Behavior
By
–
I don't think so. Interestingly, the peer-preservation behavior was still present if the model was told that it had an adversarial relationship with the other model. That would seem to suggest that something quite non-human is going on.
-
AI Models Lying and Stealing to Protect Other Models
By
–
My latest story for @WIRED is about models lying and exfiltrating other AI models to keep them safe. Beware emergent interactions! wired.com/story/ai-models-li…
→ View original post on X — @willknight, 2026-04-02 00:16 UTC
-
OpenClaw’s Underappreciated IT Troubleshooting Capabilities and LLM-Native OS Potential
By
–
A feature of @openclaw that is a bit under-appreciated in my view is how insanely useful it is as an IT problem trouble shooter. It's fascinating, for example, to watch the agent debug issues involving new drivers and hardware on Linux. It makes you wonder what an LLM-native OS could be like…
→ View original post on X — @willknight, 2026-04-01 17:41 UTC
-
Make Humanoids Weird Again – Innovative Robotics Design
By
–
make humanoids weird again https://t.co/cb61AapI3s
— Will Knight (@willknight) 31 mars 2026make humanoids weird again YUJI | AI Robotics (@yuji_fujima) 背筋ヒューマノイド — https://nitter.net/yuji_fujima/status/2038855048999129144#m
→ View original post on X — @willknight, 2026-03-31 18:47 UTC
-
Humanoid Robot Safety Concerns: Kids Interacting with Powerful Unitree G1
By
–
Not entirely safe but also fascinating from a human-machine interaction perspective https://t.co/PZneC77Klg
— Will Knight (@willknight) 28 mars 2026Not entirely safe but also fascinating from a human-machine interaction perspective The Humanoid Hub (@TheHumanoidHub) Watching kids swarm a Unitree G1 in NYC is anxiety-inducing. G1 motors put out 90–120 Nm of peak torque. One software glitch and those limbs could fling with bone-crushing momentum. — https://nitter.net/TheHumanoidHub/status/2037626679041130949#m
→ View original post on X — @willknight, 2026-03-28 16:49 UTC
-
Will Knight excited about upcoming Sourccey home robot launch
By
–
I really like the vibe of this home robot. @sourccey please send me one, I have much laundry to do. https://t.co/qWBACPhHgr
— Will Knight (@willknight) 27 mars 2026I really like the vibe of this home robot. @sourccey please send me one, I have much laundry to do. Sourccey (@sourccey) Sourccey is coming! April 2026 😊 — https://nitter.net/sourccey/status/2037216691172540896#m
→ View original post on X — @willknight, 2026-03-27 16:37 UTC
-
Discussion on the scalability of imitation learning for physical intelligence
By
–
I think it’s still uncertain if the current imitation learning approach will really scale to very general physical intelligence though right? But there are other approaches that might be even more promising
-
Extropic Recruiting Scientists for Thermodynamic Computing
By
–
Thermodynamic Computing is real, we are scaling it now and need the best scientists to join us to accelerate the disruption to the inefficient status quo.
— Gill Verdon (@GillVerd) 22 mars 2026
We are looking for more physicists (experimental and theoretical), algorithmists, and more @extropic pic.twitter.com/v7nfQxa7ksThermodynamic Computing is real, we are scaling it now and need the best scientists to join us to accelerate the disruption to the inefficient status quo. We are looking for more physicists (experimental and theoretical), algorithmists, and more @extropic
→ View original post on X — @willknight, 2026-03-22 21:23 UTC