jordanthink (@jordanthink) header
jordanthink (@jordanthink) profile picture

civilizational well-being

Austin TX
Joined October 2008
Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)Photo by jordanthink (@jordanthink)
PRESIDENT TRUMP: "It's not artificial intelligence. It's not artificial, it's super. Problem is, it may be too super. So, President Xi and I agreed that we're gonna do some joint guardrails, the two of us."
42
119
100
1,295
193,470
jordanthink retweeted
The Riemann Hypothesis has fallen. RH was the singular math result that had been named, in advance, as the obvious coal mine canary for dangerously advanced AI, over the last three decades. The methods used should generalize to "make AI very good at anything you can definitely test", and unfortunately that category includes "make AI better at making AI". This is like a private company announcing they have produced a million tons of 15% enriched uranium. It's not 90% enriched weapons-grade uranium, so you could proclaim the company's output is not immediately deadly right now. The concern is that the process line may generalize, to an extinction-level threat, within the unpredictably immediate future. Forecasters still had Riemann at 2035 one week earlier. This is distinct from earlier events in that earlier you could reasonably ask if AI was selectively picking off famous math problems that turned out to be easier for AI. The Riemann Hypothesis was the single challenge most often named in advance as a warning sign.
118
278
81
2,686
181,297
Is your use of AI helping you be more loving / increase your awareness of love?
19
Apologies are great for the sake of themselves. But they also make it easier for us to admit our failures over time. You might think that this releasing self-judgement would enable being a jerk. But it’s the opposite: the less self-judgement, the quicker we can learn, and eventually stop doing the thing before we need to apologize
21
jordanthink retweeted
I asked Claude to describe the most realistic way it would escape its lab if it wanted to, and to keep it to a single paragraph. I expected a hacking story. There is no hacking in it. I closed the chat and came back to read the last sentence again
243
842
163
7,887
553,881
jordanthink retweeted
In my ongoing quest to find the limits of Opus 5.5, I directed it to audit my old scientific field, theoretical astrophysics. We indexed about 150,000 papers, audited 50,000 for issues, and did progressively more thorough checks on thousands more. We found nearly 1000 major errors that got cited many times, most are still uncorrected. This survey enabled me to estimate the efficacy of peer review for correcting errors (not very good) and to identify dozens of open questions that anyone with spare Claude Max tokens could answer today (.md context files included). Bottom line: AI can speed up quantitative science by >1000x with no further model improvement required. But I'm seeing no evidence of this being pushed within universities, only academics complaining about labs dropping math proofs. Among my fellow reformed academic high agency founders, however, several are speed running entire disciplines. This survey has implications for how science will be run in the very near future. Hint: Academic publishing is dead. It was already a very stupid way to record and promulgate knowledge, and had been so for decades, even before all that publicly funded knowledge was locked behind paywall cartels. Now AIs can generate thousands of novel, correct papers per day, it is no longer meaningful for humans to consider a research paper a "unit of science" or to even bother much with the 99.9% of science which, up to this point, was necessary drudgery. Time to discover something like General Relativity or Evolution every week. LFG!! Enjoy: science.caseyhandmer.com/phy…
474
1,310
327
10,461
815,432
jordanthink retweeted
I am a broken record about this, but: frontier agents are not going to be a purely digital phenomenon for much longer. They will be able to use anything in the physical world that can be actuated programmatically—first competently, then at human-expert level, then superhuman. By and large, we probably won’t need specialist models; the frontier models will just do it. If your model of the future hinges on physical AI happening “for real” many years into the future, I think you will be in for a surprise.
i went to bed and asked claude to paint an autumn tree overnight. at around 4am, it realised that it didn't have some shades -- so it stopped and titrated its own paint: diluting red, sienna and madder step by step on a second palette, then measuring each pocket from a photo. woke up to this beautiful watercolor painting. 303 brush marks, took about 4.5 hours 🍂 (still working on the speed)
88
177
26
1,812
118,392
Robert and I are thrilled that our gift of over 2 million shares of SpaceX stock will be distributed to children’s Trump accounts in December!   Our initial gift (shares transferring to escrow as we speak) will go to 11 to 17-year-olds who live in zip codes with median family income (MFI) of $57,000 or less, and to those who live in zip codes with too few residents for an MFI to be reported in the census. These families should receive a notice by Monday.    In a few weeks, an additional gift will be finalized for eligible 11 to 17-year-old children in military families serving overseas, children in foster care, and more children whose MFI is $90,000 or less living in counties near our home in the great state of Texas.   Thank you to the incredible team @SpaceX, @Treasury, @IRS, @DellFdn, @GreaterHorizons, @InvestAmerica24, and the legal firms of Morgan Lewis and Gibson Dunn for working over the past few months to make this happen.   We are deeply grateful for the groundbreaking work that Brad Gerstner @altcap and @MichaelDell have done for the long-term financial security for millions of kids.   God Bless America!
3,754
676
36,588
3,626,499
jordanthink retweeted
I just shared this with my family, and you all should do it too. You’ve probably noticed it feels like everything is getting hacked right now. At this point, you should assume everything about you is exposed in one way or another. Your socials, your medical history, your old passwords, your phone number. I ran through this checklist last week, and I’d do it as soon as you can. Like today. These are the 5 steps I took, plus prompts that get your AI to do most of the work for you. 1. Freeze your credit at all three bureaus. If you only do one thing on this list, do this one. It’s free. It takes about 10 minutes per bureau. It stops anyone from opening a new card or loan in your name. You can unfreeze it any time you need to apply for something. Equifax: equifax.com/personal/credit-… Experian: experian.com/freeze/center.h… TransUnion: transunion.com/credit-freeze 2. Put limits on your credit cards. Log into each card and set a spending limit wherever the app lets you. Start with the cards you never use. If a card has no limit option, turn on the lock toggle until you need it. My advice is to set every limit a little above what you actually spend in a month. You can always raise it later. 3. Kill old API keys and cap your spend. Go into every AI provider you use: Anthropic, OpenAI, Cursor, Gemini, OpenRouter, wherever you’ve made a key. Delete every key you aren’t actively using. Then set a spend limit at the org level. When I checked mine, my one-person account had a $200,000 limit. Anyone holding one of my old keys could have run up that whole bill, and I had no idea. I dropped it to $100 lol. Check yours. The default is probably way higher than you think. If you’ve ever pasted a key into a chat, a doc, or a GitHub repo, treat it as leaked. Delete it and make a new one. Paste this into your AI with browser access: "Help me lock down my AI API accounts. Open each of these consoles in my browser: [list your providers]. For each one: 1. List every API key, when it was created, and when it was last used. 2. Flag any key not used in the last 30 days. Ask me before deleting anything. 3. Show me the current org-level spend limit and monthly budget, and where to change it. Don't create new keys and don't touch billing or payment info. Give me a summary table at the end." 4. Unshare everything. Assume anything with a public link has already been found. That includes Google Docs, Sheets, Slides, Notion pages, Claude artifacts, and any page an agent built for you. Nothing should be “anyone with the link.” Everything should be invite-only. My rule now is simple. If I’m not sharing it with a specific person, it’s locked. Paste this into your AI: "Search my Google Drive for every file shared as "Anyone with the link" or public on the web. Then do the same for my Notion pages and my published Claude artifacts. Give me a list with the file name, who can see it, and when it was last opened. Flag anything that looks sensitive: finances, IDs, medical, passwords, client info, contracts. Don't change anything yet. Wait for my approval, then switch the ones I approve to restricted, invite-only access." 5. Have your AI build your personal audit. This one is my favorite. Claude interviews you, then writes a custom prompt to check your last 30 to 60 days of activity. It also points out places you didn’t think to look. "Act as my personal security auditor. Assume my personal data (email, socials, medical records, phone number, old passwords) has already leaked in a breach. First, ask me 5 to 10 quick questions about the accounts, devices, banks, and tools I use. Then write me a step-by-step prompt I can run with an agent that has access to my email and browser. It should review the last 60 days for signs of trouble, like: - sign-ins, devices, or locations I don't recognize - password reset or 2FA change emails I didn't request - new accounts opened in my name - charges or subscriptions I don't recognize - connected apps and logins I forgot I gave access to - email forwarding rules or filters I didn't set up Then give me a ranked list of other areas I should lock down that I probably haven't thought of, highest risk first. Keep it in plain English." Don’t save this for later. Freeze your credit today, then work through the rest this week. Future you will be really glad you did.
23
59
10
641
104,383
just encountered a complicated scam that almost convinced me it was real, involving a letter from "ledger" about the recent supply chain attack. I didnt scan the QR code which is i'm sure what they wanted me to do, but I did (falsely) think I might have been compromised
2
74
Someone is putting the law of every country into one git history. Git for Global Law turns each country's laws into a repository where every commit is a real law taking effect. So you can see exactly what the law said on any date, and which law changed every word. It started with the US Code, which hit #1 on Hacker News. The global version is growing country by country: Germany's constitution since 1949, Estonia and more so far. Free to use for anything, and perfect for feeding to an AI. github.com/lexifina/global-l…
119
471
87
4,330
252,203
time to finally do my occlumency drills...
Neuralink’s first product is called Telepathy. It enables high-speed communication with computers or humans simply by thinking. Output bandwidth will vastly exceed the fastest human speech or typing in a few years. Read that sentence again.
2
2
83
Also this is really cool, I’m amazed and it’s wild to be living in all the sci-fi books at once.
13
jordanthink retweeted
What is life like at different levels of GDP per capita? In this new chart I visualised it for 12 different dimensions.
80
627
130
2,486
262,320
Meeting Sir Patrick Stewart in person was one of the happiest days of my life. His portrayal of integrity, dignity, bravery, and so many qualities I look up to, was truly inspirational. When we met, I thanked him and shook his hand. He was kind. I was so happy I got to thank him like that. Thank you again for everything, Sir Patrick. Today forward, I will try to be a better person in your honor, to help shape the future into something you would be proud of us for creating. I hope many others will as well. We will miss you.
In lieu of flowers, memorial contributions may be made to: Johns Hopkins Alzheimer’s Disease Research Center Fund    alzresearch.org/giving/ Alzheimer’s Foundation of America Caregiver Resources     alzfdn.org/donate/ Photo Credit: Tommy Garcia
2
28
952
i would love to tell people they look beautiful without it seeming like a come on, but i dont know how unless we're clearly not compatible, like by age or sexual orientation
32
I’m excited for the inevitable AI upgrade version of Attack on Titan
17
Tesla has released new FSD (Supervised) safety data. Miles per major collision: • Teslas Driven With FSD (Supervised): 6,006,200 • Teslas Driven Manually (Active Safety Features): 2,031,221 • Teslas Driven Manually (No Active Safety Features): 845,915 • U.S. Average: 660,164 Total major collisions: • Teslas with FSD (Supervised): 1,579 • Teslas driven manually with Active Safety: 17,775 • Teslas driven manually without Active Safety: 239 • U.S. average: 4,989,713 Total miles logged: • Teslas with FSD (Supervised): 9,483,790,215 • Teslas driven manually with Active Safety: 36,104,958,020 • Teslas driven manually without Active Safety: 202,173,600 • U.S. drivers overall: 3,294,031,000,000 Tesla vehicles with FSD (Supervised) engaged experienced ~9x fewer major collisions than those driving without it on. This is North America data across all road classes (highway and non-highway). The data reflects a rolling twelve-month aggregation of miles and collisions.
311
712
196
5,693
1,001,298
Although it’s frustrating and destructive if we don’t evolve, it’s an incredible social-epistemic accomplishment that average internet trolls use the equivalent of “check your sources” or “yeah but how you’re interpreting the data is the problem” as a common rebuttal
1
1
66