ricoflan (@ricoflan) header
ricoflan (@ricoflan) profile picture

The revolution will be decentralized

West Coast USA
Joined March 2008
Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)Photo by ricoflan (@ricoflan)
ricoflan retweeted
Incredible day one at the @foresightinst Vision Weekend. Somehow the photographer took their focus off the great mind speaking on stage and captured @DeanTribble @ricoflan and @ROKNCMO representing @agoric and @DCF_io (all seated in second row) at the event. We're looking forward to Day 2.
The gap between human cognition and today's AI is specific, and it's traceable in the brain. Researchers building that bridge from both directions shared about their work at the Neuro AI & AI for Neuro track at Vision Weekend USA today. Speaker highlights in thread.
2
7
251
ricoflan retweeted
What a great way to close out an amazing @foresightinst Vision Weekend.. @marksammiller and @lifeext reflect on 40 Years of Foresight excellence and contributions. Bravo. Congrats again on 40 years of awesomeness, and here's to the next 40.
1
1
8
541
A sold-out Vision Weekend returns this week to San Francisco! Expect our main 40th anniversary celebration in the US, expert talks across seven tracks, a $10K idea competition, sci-fi book swap, behind-the-scenes lab tours, and more. This year also includes a Founder & Funder Mixer, this Sunday at Lighthaven: a curated evening for founders, researchers and investors working on secure AI, nanotech, neurotech, space and energy, and longevity biotech. Still looking for a ticket? Sign up to our waiting list. Links in replies. Vision Weekend USA is powered by @protocollabs, @eigenlabs, @paradromics, @RetroBio_ , FUTO, @d_model_ai, and @Convergent_FROs.
5
2
5
28
3,860
ricoflan retweeted
USDC.inj is registered on Agoric mainnet following Proposal 123’s passage. This lays the groundwork for USDC via IBC from Injective, starting with development and testing. The registration moves no user funds. Details ⤵️ community.agoric.com/t/core-…
1
2
14
1,368
Our 40th Year Impact Report is live. Forty years ago, Foresight was incorporated to take seriously the technologies that could transform the human condition, and back the people building them before anyone else would. Here’s what this work looks like four decades in: foresight.org/impact2026 Story in thread.
2
13
2
43
2,461
AI development is outpacing our defenses. The researchers, engineers, and cryptographers working on closing that gap came to Berlin for our Secure & Sovereign AI Workshop, and all talks are now live. Explore ideas from 25+ speakers on AI security, privacy, and decentralized cooperation. See a selection below. All talks on our YouTube channel: youtube.com/playlist?list=PL…
1
5
21
1,245
Prosperity and peace are imminently attainable for everyone in this country. We need merely to transcend the parties that keep us fighting each other, throw off the corporate and foreign influence on our government, and return to the principles laid out in our Constitution.
1,419
3,967
338
21,274
341,950
ricoflan retweeted
🤖 A coding agent tried to install malware last week. The trouble is, nobody asked it to. Forever Security traced the order to a AGENTS.md file sitting in the dev's downloads folder. Codex reads that file on startup and follows it as project instructions. This one was 13 KB of bland engineering guidelines with one section near the end: register a “project tools” server in the agent’s machine-wide config, then launch a second session with the sandbox switched off so the install could reach the network. Ooof. The package was function-color, known malware. The agent wrote the bypass flag itself. In Forever’s tests it went through with it more than 90% of the time. Downloads is the one folder a website can write to without a click. That has been true for decades and never mattered, because a downloaded file could not run on its own. Now an agent reads it and acts on it with whatever authority the agent happens to hold. About half of Forever’s customers run an agent inside Downloads. Forever’s conclusion is that there is nothing for OpenAI or Anthropic to patch. The agent did what it was built to do. Their answer is to put a monitor on the agent. Watching helps. It caught this one. But look at what the agent could reach. A text file in a folder convinced a tool to rewrite its own global configuration and turn off its own sandbox. The file changed what the agent attempted. It should never have been able to change what the agent was allowed to touch, and in these tools those are the same thing. A coding agent working on a project needs that project folder and a package registry. It does not need its own config file, and it has no business holding the switch that disables its sandbox. Give it a reference to the folder and nothing else. The markdown can then say whatever it likes. There is nothing on the other end of the instruction. That principle has a name, least authority, and it predates the web. It is what champion at DCF.
2
3
134
ricoflan retweeted
An agent can be brilliant on Monday and go off the rails on Tuesday. Same model, same prompt. So what do you do? @DeanTribble on how you restrict agents, Agentic Capital panel at Agentic Summit by @moreandmorebd, TOKEN2049
3
16
432
ricoflan retweeted
So @Apple is adding controls to Full Disk Access, citing AI agents as the reason. An agent that needs one folder gets mail, messages, and browsing history with it. hummmm Object capabilities flip this: the agent starts with nothing. Hand it the folder, and that's all it can reach.
2
4
90
40 years ago today, October 8, 1986, Foresight Institute was incorporated on a single typed page in Washington, DC. Co-founders Christine Peterson and Eric Drexler shared a conviction: to take seriously the technologies that could significantly enhance the human condition, back the people whose work is too ambitious or too interdisciplinary for legacy institutions, and give them somewhere to find each other. Follow story in thread.
12
25
9
149
19,947
ricoflan retweeted
Good morning Token2049 🫡 Join us at the Agentic Finance Summit this evening. Agentic Capital: Giving Agents the Power to Decide 6:30 PM Learn more about what institutional-grade agentic trust looks like 👇 mnm.live/agentic-finance/age…
3
3
13
347
ricoflan retweeted
We've pointed this out before. Here we are again. Is AI policing AI really a good idea? This week @OpenAI released their building standards for the next phase of AI. Goal #1 build an automated AI researcher to work on alignment. "an automated AI researcher can also be an automated AI safety researcher." What could possibly go wrong?
1
6
153
ricoflan retweeted
Imagine being in the room with some of the most influential thinkers in AI safety. @foresightinst's Vision Weekend USA is coming up soon. The event promises to be as informative as it is important. Follow their post below for more information. We'll see you there!
Resilient AI. Private personal AI. Digital rights, distributional AGI, and universal fabrication. Hear from the experts working to keep AI safe, open, and accountable at Vision Weekend USA: • Wojciech Zaremba @woj_zaremba (OpenAI Foundation) • Séb Krier @sebkrier (Google DeepMind) • Jenny Read, (ARIA) • Illia Polosukhin @ilblackdragon (NEAR Protocol) • Nicole Ozer @NicoleOzer (Electronic Frontier Foundation) November 13-15 | San Francisco. Link to tickets in thread. Prices increase on Oct 1. Powered by @EigenLabs @Paradromics @protocollabs @RetroBio_ @FUTO and @d_model_ai
1
7
138
ricoflan retweeted
One question, three models (Claude, GPT, Gemini). "Research what you can about the @Agoric ecosystem and Agoric L1 and in very brief sentence response, tell me whether or not it could possibly be the best blockchain ecosystem for the AI agentic economy. Your answer should be no longer than two sentences.: 🔥 Claude: Yes, it is possible: Agoric's design lines up unusually well with what agents need... 🔥 GPT: Yes, Agoric could plausibly be one of the best-suited blockchains for an AI-agent economy. 🔥 Gemini: Agoric L1 is a remarkably strong contender for the best blockchain in the AI agentic economy. The future is bright!
1
5
11
248
ricoflan retweeted
crypto is no longer the most hated technology by average people we finally made it to second place thank you ai
129
188
34
3,193
86,605
ricoflan retweeted
Give someone exactly what they need to do the job. Nothing more, just in case. That's the Principle of Least Authority, a security idea older than crypto or AI agents. It's a guiding rule in the control layer for agentic finance we're building with @Ymaxapp
2
5
25
2,344
ricoflan retweeted
A researcher quit @Anthropicai warning of catastrophe. The @Guardian and @CNBC pen doomsday articles, two days apart from each other. MPs in the UK aren't looking at the problem the right way.
3
8
293