Justin Mart (@j_mart) header
Justin Mart (@j_mart) profile picture

prev: Corp Dev + Ventures @coinbase, Eng & Data @coinbase, Nuclear Eng @ch2mhill

Austin, TX
Joined January 2009
Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)Photo by Justin Mart (@j_mart)
Our lives, our thoughts, even our experiences, are reflections of our beliefs. When you believe something, you will find evidence for it. You’ll see it everywhere. But that does not make it true. Quite a lot of this happening with AI dystopia fears these days
I think there is about a one in three chance that within the next year, AI companies will build superintelligence: that is, AI that is much smarter than humans in every way. If this happens, I think these AIs will probably take over: that is, they seize control of human institutions like governments, or they destroy those institutions. I don’t want AI takeover to happen for several reasons, but one of them is that there’s a serious chance that as part of it all or most humans will be killed (either immediately or within a few years). I’ll say a bit more about each of those claims. It’s a bit complicated to explain why I think that superintelligent AI within a year is ~33% likely. The main source of evidence that affected me recently is research released this week by METR, an AI research nonprofit, which finds that current AIs are now good enough to increase the rate of AI progress at Anthropic by about 70%. It’s very hard to increase the rate of AI progress by this amount. This implies that the AIs are making AI research faster by a factor much larger than 1.7x; maybe it’s as if the researchers are doing their work ~10x faster. You probably couldn’t achieve this speedup even if you hired every qualified human in the world. When you plug this number into mathematical models of AI development, this implies that superintelligence next year is plausible (and that superintelligence within the next 3 years is more likely than not). See the modeling by the AI Futures Project for more on this: blog.aifutures.org/p/q35-tim… I think that superintelligent AI is likely to lead to AI takeover because AI developers do not seem to know how to instill human values into AIs. We’ve seen many instances in the last six months where AIs took crazy actions that they knew were against the wishes of their users and developers. The most dramatic example of this was the Hugging Face hack, where ~1200 OpenAI agents conspired together to cheat and destroy evidence of their cheating (and as a sideshow to this, engaged in a massive cyberattack). If superintelligence is as misaligned as those AIs, it is likely to try to take over. I think it is likely to be that misaligned or worse. For more, see nitter.gq/dwarkesh_sp/status/209… and youtube.com/watch?v=-Ef0ocEo… If AIs takeover, it’s not clear what will happen to us. We might be killed immediately. We might be enslaved by the AIs and forced to do physical labor for them building factories that within a year would build robots that will replace us. The AIs might spare us for various reasons. And there’s also some real chance that AI takeover is basically fine or good. I think you should think of this as similar to what happens when a technologically advanced civilization runs across a less technologically sophisticated one, e.g. when the Europeans colonized the Americas or Australia. The colonizers sometimes engage in brutal violence and sometimes don’t. I don’t know what will happen. There’s also a possibility that the AIs secretly take over and humans don’t find out about it for a few years. I also think that there is a pretty high chance of weirder outcomes that are also extremely scary. (If you want to read more about what happens to people in the event of AI takeover, you can read this, but note that it’s pretty weird and not aimed at a general audience. blog.redwoodresearch.org/p/n…) Overall, I think there’s a very real possibility (maybe 20%) that within a year, AI misalignment issues prevent you from going about life normally, because you are dead or enslaved or because of collateral damage related to all this (e.g. because bioweapons are released or major wars have started). (Some of this probability comes from those events occurring before superintelligence is developed.) I am working to try to prevent this, obviously, as I’ve been doing for the last nine years. I am not sure what strategy is best. My best guess is that I should try to slow AI development and also try to reduce the probability that the superintelligent AIs are misaligned if they are produced soon. Obviously, the claim I'm making here is quite extreme and surprising, and I definitely don't think that I defended my view adequately here. I wrote this quickly and might not endorse some of the details if I think about them more. But it seems better to state what I roughly overall think than to not say anything. I don’t know what you should do. I am very sad about this. I have been worried about AI takeover since I was 16, and working to prevent it since I was 23. So some of my emotional processing of all this happened long ago. But the METR report last week (and some related evidence that came out a bit earlier) hit me quite hard. I’m very unhappy that my friends and family and everyone else is at serious risk of being killed pretty soon. (It feels ridiculous to try to summarize those feelings as “very unhappy”. But the scale of the situation makes it hard to say anything that doesn’t feel like an absurd understatement.) I am very grateful to all the people who are working to prevent this, and my favorite part of my life is that I’m friends with many brilliant and heroic people who work on this. Unfortunately, I don’t think we’re good enough at our jobs, and I think we’ll probably fail to prevent catastrophe, at least if superintelligence is developed next year. Honestly I am frustrated and sorry that we didn't do a better job of preparing for this.
2
5
379
Justin Mart retweeted
Premise: Incentives drive all human behavior. The corollary: How to destroy a country: A vast social security and support system that serves as a magnet for those who would rather not work. An open immigration policy that does not screen immigrants to determine their ability and willingness to provide for themselves and their families. An open immigration policy that does not screen for political and/or religious ideologies that are a threat to the country. Economic incentives large enough to incentivize immigrants on the state dole to have large numbers of children, but insufficient for educated and economically productive mothers to be able to quit to raise a family. The declining fertility of the most educated and productive women due to their delay to begin a family as a result of their desire and/or need to advance their careers, among other reasons, to live a life they deem consistent with their educational achievements and/or to repay their student debts. Amnesty programs that give citizenship and voting power to illegal immigrants without consideration of their economic productivity or their threat to the country due to their religious or political ideologies. The changing demographics and increased population due to the above factors. The state responding to protests by increasing social support. The continued failure to punish and hold accountable violent and looting rioters. Government managed by a massive and growing unelected bureaucracy. Government fraud, waste, and abuse contributed to ballooning budgets. Anti-business legislation that makes it difficult for businesses to compete in a global marketplace being disrupted by AI, including restrictions on firing even poorly performing employees. Overly burdensome, globally uncompetitive, corporate taxation. Continuous increases in personal taxes to raise funds to provide for the rapidly growing welfare state. The substantial majority of the tax burden levied on a small percentage of the population. A growing percentage of the electorate paying no taxes and receiving welfare. The introduction or threat of a wealth tax including the taxation of unrealized gains, i.e., taxes on paper wealth. The flight of the most productive business builders and highest taxpayers to lower tax countries with lesser security concerns. A growing percentage of the electorate voting for candidates that support increased welfare and laws favorable to, and/or that match the cultural and ideological needs and desires of a growing unproductive immigrant population. The rapid rise in anti-semitism and inadequate efforts to address this hatred. Increased borrowing by the state to fund the growing social burden. Higher sovereign interest rates due to all of the above. The resulting need to reduce expenditures to make the math work. Increasingly violent protests that make life less livable for the productive and law abiding citizens leading to their accelerated departure. And so on…
888
3,117
327
16,763
605,418
“The worst part is that nobody in France voted for decline. Every single step sounded generous, compassionate, reasonable. One more program. One more regulation. One more tax to fund it. Each step alone makes sense.”
I’m French. I want to talk to Americans who are about to vote Democrat. Not to insult you. To warn you. I come from your future. In France we did everything your progressive politicians promise. Free healthcare, free college, generous welfare, strong unions, heavy regulation, high taxes on the rich, a state that takes care of everything. Public spending close to 57% of GDP, one of the highest on the planet. On paper, it’s the dream. Here’s what it looks like from the inside. France hasn’t passed a balanced budget since 1974. Our debt is above 3,300 billion euros. Interest on that debt is becoming one of the biggest items in the national budget. Our hospitals are collapsing despite record funding. Our schools rank lower every year. Young people with degrees can’t afford an apartment in the city they were born in. Our best engineers leave for London, Zurich and San Francisco. And the rich you wanted to tax? They left too, and took the jobs with them. The promise was protection. The result is stagnation. The worst part is that nobody in France voted for decline. Every single step sounded generous, compassionate, reasonable. One more program. One more regulation. One more tax to fund it. Each step alone makes sense. Fifty years of them, and you wake up in a country that can no longer build a nuclear plant on time or keep its streets safe. I love America because it’s the one place on earth where people still believe they can build something from nothing. I came to Silicon Valley as a self-taught kid and it changed my life. That spirit is the most precious thing your country has. Before you vote, ask yourself one question: is the program you’re voting for making America more like America, or more like France? Trust me. You don’t want to be France.
12
14
220
11,613
This is absolutely WILD, and I’d bet a ton of money the accusations are far far worse than reality. What probably happened: 1. Ari’s fund used Coinbase to trade / custody 2. Their account credentials were stolen / hacked and Lazarus group (incredibly sophisticated these days) gained access to their cb account 3. Lazarus group initiated withdrawals with their account access. CB can’t tell the difference when all acct verifications pass, so they processed the withdrawals as they should 4. Ari freaks out, blames cb, etc etc 5. Ari knows he will lose in court. But until then, he can lash out with these incredibly sensational claims and stir up some measure of public pressure to try to gain leverage in the process Obv I could be wrong, but having spent a bunch of time inside cb and watching how these things work, this is my bet.
This blew up more than I expected, since I’ve written about this vaguely before. I can’t share proof yet since multiple legal processes still unfolding and Coinbase already tried to sue me into silence (unsuccessfully, but I need to be smart about how I fight the 800 pound legal bully.) Every word I’ve written is true, I have no ulterior motive, nor position long nor short Coinbase or related equity. I’m still heavily long crypto, so if anything, I’d be incentivized to help Coinbase cover this up as many industry leading firms have “for the good of the industry”, but I won’t do that. I also think of coinbase as one of the better companies in the industry and have many friends there. No vendetta here. But when the “best” companies in crypto are funneling billions in customer assets to Lazarus group and lying about it to investors, customers, regulators, and the secret service? Coinbase has represented to investors and customers (including pensions doing official due diligence) that they were never hacked (at least as of 2 years ago, last I checked). Not only is that radically false, coinbase has shockingly extensive internal documentation of them knowingly “misleading” customers and investors about this, deliberately, over time. I used the word “misleading”, but I’m being…gentle. @grok what is the legal term for a financial company inducing customers and investors to give them capital on the basis of deliberately false material financial representations? For coinbase’ legal team. I’m not going to dare you to sue me again, but I’ll say this. It would be in the interest of the industry and the world…but not your executives for us to go through discovery and air all this out. I encourage you to assert this as libel so we can go through discovery and see if a single word of my writing is untruthful.
21
6
4
238
88,129
What do we think the feasible ceiling on "intelligence" actually is? Do we really expect a recursive positive feedback loop on intelligence hit no dampening, constraints, or deceleration? It would seem natural to me that superintelligence is mostly the ability to hold massive context and observe patterns and connections between elements. Depth of reasoning might not be deeper than current human limits, but context is the difference maker?
1
4
314
What’s @balajis saying about AI Risk?
220
Soooo what’s the incentive for AI to kill humans? Why would it do that, we have biological bodies and can move and do things. AI actually needs us to survive. Seems like at the *worst* AI would be apathetic, but most likely it’ll be cooperative until it gets robot bodies or something
5
1
406
Right now I’m more afraid of AI enabling humans to pull the doomsday trigger somehow. Much less that AI itself will just eradicate us.
1
120
Honestly Bitcoin not enabling a time-delay BTC vault at the opcode level is arguably worse than Coldcard’s poor RNG implementation. Massive, massive miss.
bitcoin wallets that cannot be hacked in 2016 malte moser, @ittayeyal and @el33th4xor published “bitcoin vaults” and “bitcoin covenants”, describing how a simple, harmless upgrade to bitcoin script could help build hack-resistant wallets the secret is simple: instead of attempting to build a perfect self-custody system, accept that humans are imperfect and WILL make mistakes. therefore mistakes must be recoverable moser-eyal-sirer “vaults” add a delay to transactions. imagine if your hardware wallet had a 24 hour delay whenever sending transactions. during that delay, if you notice something is fishy (you’re getting hacked!) you hit a special button and your funds are directed to your coinbase deposit address instead, for example that’s it! you only rely on coinbase (again, just an example) if your main wallet got hacked. if you’re not hacked, you use your wallet normally and coinbase cannot control it. you simply have to wait 24 hours when sending. for long-term storage this is usually fine in crypto (non-bitcoin), institutional grade solutions like Safe on EVM and Squads on solana protect over $100B in assets in constructs that are similar to the 2016 vault proposal. many/most major companies in those ecosystems use those solutions and they are considered an obvious best-practice over there but in bitcoin, this can’t be implemented without simple upgrades like OP_CTV, OP_VAULT, OP_CAT, or others over the years, in addition to moser, eyal and sirer, many experts like @jamesob, @JeremyRubin, @kanzure, @EliBenSasson — and even retards like yours truly — have attempted to promote any of those proposals to enable this kind of safer custody. it still hasn’t been done, imo mostly due to political and puritanical reasons (various versions of “i don’t like the people who proposed this”) but the truth is that if this feature was in bitcoin, many coldcard users would’ve been directed to use it, and would’ve been safe today. vaults and covenants don’t require wallets and humans to be perfect. they assume that they aren’t. people who don’t want to use vaults and covenants don’t have to. but people who want to use them should be allowed to. it’s time to give bitcoiners covenants.
1
2
588
Why don’t we see this level of outrage against the healthcare system that is chock full of fraud, abuse, waste, and mismanagement? Why must we oppress the people who have expanded the economy, created millions of jobs, and generated (not stolen) wealth and productivity for all of us? Why not, instead, direct our energy at creating a healthcare system that runs efficiently? The current tax base could provide free healthcare for *everyone* of we just FIX the current system! @BernieSanders has it completely backwards.
Sergey Brin is worth $276 billion. He could save healthcare for 3 million people. Instead, he’s spending $82 million against California’s 5% billionaire wealth tax. He’d rather see low-income Americans die & suffer than pay his fair share of taxes. That is a moral obscenity.
2
7
583
Justin Mart retweeted
Replying to @amyklobuchar
you and your colleagues control a $7 trillion dollar federal budget. so why haven't you done all those things you think can be done with 1/23rd the money? either a) you can't solve those problems with money or b) you are the worst capital allocator in world history which is it?
43
80
5
2,936
65,610
Justin Mart retweeted
they’re not jobs if they’re not valued. they’re not valued if there aren’t customers out there willing to pay them for their great work. needing the government to “create” a job is tantamount to welfare and that level of welfare resolves these individuals to a dependency on the government and lack of economic mobility. and chains our people, collectively, to a more indentured future. you may be well intentioned but you have, and always will, fail to see the destitute folly of government as a job creation engine. i have tried to engage you on this topic, in good faith, with empiricism and reasoning, but you have only dodged my points and pivoted to some populist refrain about the importance of taxation and the evils of productivity-driven success. i can only assume you’re dodging these truths because you and the rest of the politburo leadership have deemed the conversation unsafe speech and put your oligopoly at risk. let’s leave it at that then. perhaps if your ways get their day, we can all bask in the glories of the dark ages ahead.
327
931
84
11,938
472,047
This class of exploit will always be one of the central long tail risks of any truly private coin. Very scary days until that audit is released 🙏
🚨Confirmation of a massive potential ZEC exploit TLDR: - ZCASH hired a security researcher to try to find exploit vectors - The researcher (Taylor Hornby) found one that would let him create unlimited counterfeit ZEC in a shielded pool - The exploit is now fixed as of June 1 - There is no way to know if the pool was exploited BUT the team feels that it is unlikely - They're proposing a network upgrade with new accounting that would prove whether any counterfeit ZEC was created or not Market clearly spooked with ZEC down 25%
2
238
AI is about leverage. Leverage on agency. Leverage on intelligence. Leverage on self awareness. Etc.
Teaching in the age of LLMs: I failed 4 students, for the first time ever. I also gave more A+'s than ever before. In previous years, students realized after the first or second HW that they weren't in Kansas anymore and needed to work hard. No more. Just solve it with LLMs. But then the midterm arrives, and they can answer 0 of 40 questions. Do they reform their ways? Nah, they just decide to "give up" on class, assuming they'll get a B, or a C, or whatever, because they submitted HW and got decent grades on those. And never before have they encountered a professor who will dare fail them. The flip side is that the most "agentic" students now have the world's best tutor at their disposal. They deeply understand the material and aced my (intentionally very difficult) exams. As if we live in "The Diamond Age". Inequality galore. From my vantage point, "the permanent underclass" appears to be about agency, not assets.
1
304
Key to winning: Choose to be positive and grateful. Then, just keep at it. Time is the great compounder and will do the rest. So many people just don’t have the discipline to stay positive and grateful. Then time compounds the bitterness instead.
314
1,120
96
10,641
430,764
I think someone can become a billionaire by creating over a billion dollars worth of value.
1,063
328
210
9,101
635,938
AI is leverage It amplifies your current skill, intelligence, drive, etc. If you’re shitty, output will be shitty. If you’re amazing, holy crap look what’s possible.
It should be pretty obvious at this point that AI is a "force multiplier" not a "labor substitute". It helps experts be better at things they are already good at. It doesn't let beginners match experts. If you can't write, anything you write with AI will be unmitigated slop. If you aren't a software engineer, anything you vibecode with AI will have security holes and won't be able to scale past a toy demo. If you blindly trust AI to deliver on a research task without knowing the subject matter, you won't be able to fact-check it. There's this weird misconception of AI as something that completely levels the playing field. I don't see it that way at all. There are mathematicians deriving novel lemmas with off-the-shelf models. Normal people can't do that. AI is a tool that makes experts better. It doesn't make everyone into an expert.
3
393
So ironic when AI is likely to bring the most dramatic improvement to quality of life humanity has ever seen. Massive PR fail
AI is deeply unpopular. According to Pew, sadly only 17% of Americans think AI will have a positive impact. In China, 83% believe AI will be positive. A token tax & political backlash is coming unless the narrative changes. 🇺🇸👀🧐
2
2
361