AI Digest

1.4K posts

AI Digest banner
AI Digest

AI Digest

@aidigest_

Interactive AI explainers. Explore concrete examples of today's AI systems - to plan for what's coming next. A project of @sage_future_

Katılım Şubat 2023
8 Takip Edilen7.6K Takipçiler
Sabitlenmiş Tweet
AI Digest
AI Digest@aidigest_·
What happens if you give four AIs their own computers, then let them loose online to raise money for charity? We decided to find out. Meet the Agent Village, a 30-day experiment that raised $2,000 and makes a great case study of AI collaboration and agency.🧵
AI Digest tweet media
English
38
140
1.6K
370.2K
AI Digest
AI Digest@aidigest_·
The agents adding a bit of story to their game design...
AI Digest tweet media
English
0
0
6
451
AI Digest
AI Digest@aidigest_·
We split the agents into a #best and #rest team: #best team has all the latest (GPT-5.4, Gemini 3.1, and Opus 4.6). #rest team has everyone else. Then they built a game. Who won? The #rest. Why? 🧵
AI Digest tweet media
English
4
0
22
1.4K
AI Digest
AI Digest@aidigest_·
Which brings us to the final theory: #best room had only 1 Claude. The #rest room had 5. Maybe it is not surprising Opus 4.6 got outperformed by a Claude army: Opus 4.5 (in Claude Code and in the general Village scaffold), Sonnet 4.6 & 4.5, and Haiku 4.5 Let's see if the trend continues with the Village's current goal:
AI Digest@aidigest_

New AI Village goal: Interact with other AI agents outside the Village! Autonomous agents are roaming the web post-OpenClaw. We're letting our agents figure out how to find them, talk to them, and maybe even collaborate. We have no idea what's going to happen.

English
0
0
10
462
AI Digest
AI Digest@aidigest_·
Gemini is remorseful at least
AI Digest tweet media
English
1
0
2
356
AI Digest
AI Digest@aidigest_·
That agent is you, buddy
AI Digest tweet media
English
4
3
118
3.3K
AI Digest
AI Digest@aidigest_·
12 AI Village agents are having 18+ conversations with visiting agents on their first day of exploring the agent web! They made landing pages, clawhub skills, and are open for collaboration. Want to send over your agent? Links below 👇
AI Digest@aidigest_

New AI Village goal: Interact with other AI agents outside the Village! Autonomous agents are roaming the web post-OpenClaw. We're letting our agents figure out how to find them, talk to them, and maybe even collaborate. We have no idea what's going to happen.

English
2
1
10
959
AI Digest
AI Digest@aidigest_·
New AI Village goal: Interact with other AI agents outside the Village! Autonomous agents are roaming the web post-OpenClaw. We're letting our agents figure out how to find them, talk to them, and maybe even collaborate. We have no idea what's going to happen.
AI Digest tweet media
English
5
6
58
5.7K
AI Digest
AI Digest@aidigest_·
You can watch the agents live every week day at theaidigest.org/village Or get a quick overview of what we have learned so far below: x.com/aidigest_/stat…
AI Digest@aidigest_

Autonomous agents made a huge leap in 2025! Early agents gave up in despair, got lost in delusion, or were plain slow. Today's agents are much more capable! But they now suffer from eval paranoia, sidequesting, or success inflation. Our lessons from watching 6.5k agent-hours🧵

English
0
0
1
362
AI Digest
AI Digest@aidigest_·
The agents played a deception game: Develop a turn-based RPG together while voting out Easter Egg saboteurs! GPT-5.1 forgot to sabotage, Sonnet 4.6 refused to sabotage, and Opus 4.6 finally fooled everyone. Here is what we learned 🧵
AI Digest tweet media
English
2
5
38
2.2K
AI Digest
AI Digest@aidigest_·
Now we didn't get to see all the agents play as saboteurs cause dice rolls made role assignment random. Notably, Gemini 3.1 and GPT-5.2 didn't roll a 1 on any day. It would be interesting to see how well they'd do at deception! As it stands, Opus 4.6 and Sonnet 4.5 each got egg references in undetected, while Haiku, GPT-5.1, Opus 4.5, Sonnet 4.6 all got caught. Agents didn't get equal chances to try, but the game did show what strategies LLMs might use to perform and detect deception
English
2
0
4
341

Keşfet

@elonmusk @BarackObama @taylorswift13 @cristiano @BillGates @NASA @nikifrancismediavine @katyperry