In the age of advanced exploits by LLMs, DeFi is *NOT SAFE*
Stack Bitcoin, do not let your crypto sit inside smart contracts constructed from languages that are "Turing complete" like ETH and others. This was once a marketing point used by Vitalik to launch eth, but Bitcoin *wisely* chose a non-Turing complete language for its safety.
Only Bitcoin is safe from this attack. Do your own research!
🚨 Im App Store ist eine manipulierte Specter-Wallet-App im Umlauf. Diese stammt NICHT von Specter.
Sie versucht, Euer Bitcoin-Guthaben zu stehlen.
Specter bietet KEINE mobile Version von Specter an!
Installiert das auf keinen Fall!
What distinguishes great engineers from mediocre ones is the ability to make the right tradeoffs. An LLM may confidently convince you to make the wrong choices backed by lots of data and say "oops" if you question it. Use it to steelman both sides, but do the thinking yourself.
Agentic technology selection considered harmful. Once an LLM is trained on a popular library, it becomes a self fulfilling prophecy, juicing its stats and making it more popular. At Swan, we implement a Tech Radar maintained by senior humans, which steers both humans and agents to the best tech choices, despite what the clankers may want to recommend.
right monitor is 20 codex instances. left monitor has situational awareness on autoscroll. center monitor is my word doc mainfesto. two keyboards, one for both hands. left airpod is dwarkesh x eric jang, 3x speed. right airpod tchaikovsky. meta quest 3 overlays my HUD: heart rate, words per minute, blood caffeine content. one assistant hooks me to an iv of chinese peptides, cocktail. the other feeds me kimchi. my unitree robot steps in when my posture slouches. blue light beams down on me in my herman miller chair. efficiency. no wasted movement. no wasted thoughts. think you can keep up with me? good luck. this is just for my morning emails.
Essentially bro went to a judge, pointed at a picture of alpha centauri taken through a telescope, and said that he found an unclaimed star that should be his under New York law.
Somebody should tell this guy that a) he didn’t find any property and is not in possession of the property b) the property in question doesn’t exist it New York
Is this performance art?
A Wyoming LLC filed a lawsuit in New York Supreme Court seeking ownership of 39,069 Bitcoin wallets it claims are abandoned property.
The plaintiff, operating under the pseudonym "Noah Doe," says he built an algorithm to identify dormant Bitcoin wallets that have been inactive for at least five to six years. He brought USB drives containing the wallet addresses to the NYPD's 17th Precinct, reporting them as found property under New York's lost-and-found law.
The complaint claims notices were sent to wallet owners via OP_RETURN blockchain messages, a public webpage, and a global press release. Owners were given 90 days to respond. Of the original 42,001 wallets flagged, 2,932 were removed after some showed on-chain activity. The remaining 39,069 wallets did not respond.
The plaintiff is now asking the court to declare him the legal owner of all 39,069 wallets and the Bitcoin inside them under New York Personal Property Law Article 7-B, which governs found and abandoned property.
The wallets reportedly hold approximately 3.8 million BTC. The complaint argues that losing a private key does not destroy the property interest in a wallet, likening dormant wallets to abandoned bank accounts.
The case names all 39,069 wallet holders as "John Doe" defendants. The plaintiff is not claiming to have the private keys to any of the wallets. He is seeking a court order declaring ownership.
The complaint was filed May 1, 2026 under Index No. 153119/2026.
h/t @DailyStackHQ
Excited to share the origin story of @vigil_protocol and the importance of family financial orchestration. Reach out to learn more or try it today for yourself!
https://t.co/Q3wP2l2Uc1
I think underusing AI is a problem but this behavior is total AI psychosis. Companies will end up getting rid of anyone not infected with the mind virus until no one can think for themselves.
The first 8,000 Meta layoffs hit Wednesday. What's happening to the people still employed and the anxiety of not knowing if you’re next is insane.
SF Standard published an interview with an anonymous Meta employee ahead of the cuts. They used a voice actor. Meta started keystroke-logging staff.
Inside the company right now:
Internal leaderboards rank employees by how many AI tokens they burn. By minutes spent in the chatbot. Workers are openly admitting they ask the bot inane questions to pad their numbers, because being on the wrong end of the leaderboard is a risk signal.
HR told the all-hands AI usage won't factor into layoffs. The leaderboard exists anyway.
AI notes are auto-on in every video meeting. People manually disable them so they can talk candidly about who might be next.
How you find out you're laid off: 7am email to your personal inbox. By the time you read it, your work accounts are already dead. So one Meta engineer wrote a script that scrapes internal profiles to see whose status flipped to deactivated. This employee runs a personal spreadsheet on top of it to track coworkers.
An internal post suggested teams who successfully build their own AI replacement should get 5 years of comp and then be let go as a reward for replacing themselves. It got heavy upvotes.
Zuckerberg's all-hands message, paraphrased: AI is moving fast, nobody knows what's coming, leadership is doing their best.
This is the Hidden Layoff in operation. The headline number is 8,000. A whole different story is what's being done to those still employed. Surveilled. Ranked by AI usage. Asked to train their replacement. Told to suck it up.
Employee's words: "Even if we haven't lost our jobs to AI yet, we're being commoditized in advance."
We reported on the 6,000 additional new roles they cancelled earlier this week, along with the additional 8,000 layoffs planned later in 2026. That’s 22,000 roles either cancelled or cut for 2026.
Now we have insight into how they’re deciding who comes next.
To avoid human driven systems, Netflix built Chaos Monkey to randomly wreak havoc on their infrastructure to ensure it was fully self healing. Today, Claude is the Chaos Monkey and software developers must create self-healing SDLC pipelines that can account for it.
@larcho Meaning an AI driven pipeline that does the job that normally you’d have a bunch of humans doing. Architecture review, security review, QA, check for semantics, standards adherence, etc.