★ Adam LG Ring ★ BSc - IBM

@iamdevloper > What can go wrong? Narrator: As it turns out, just about everything.

@araseb_ Claude & Me > Claude solo

You’re in a tech interview and they ask you: “Why should we hire you when Codex can write code?” What’s your answer?

@droidbuilds dollar maxxing !

"mom, how did we get so poor?" "your father had Claude Max, ChatGPT Pro, Cursor Pro and shipped absolutely nothing"

@himes369 clones are not that useful since they copy something that exists and is legit better

"AI artık oyun yapıyor, sektör bitti" diyenlere de "AI çöp prototip üretiyor, ciddiye almayın" diyenlere de: ikiniz de yanlış yeri tutmuşsunuz. bir çocuk claude fable 5'e birkaç cümle yazdı, tarayıcıda low-poly bir gta klonu çıktı. polis kovalıyor, araba çarpıyor. evet gerçek. ve evet, three.js iskeletinden öteye geçmiyor. ama olay prototipin kalitesi değil. olay şu: prototipe varma maliyeti sıfıra yaklaştı, oyunu bitirme maliyeti aynı kaldı. > fikir, çalışan prototip: eskiden 3 ay, şimdi 10 dakika > prototip, bitmiş oyun: hala 6-18 ay > balans, loop, optimize, retention: AI'ın eli değmiyor yani darboğaz yer değiştirdi. artık "yapabilir misin" değil, "yapmaya değer mi, neyi yapacaksın" sorusu kaldı geriye. ucuzlayan şey üretim değildi. ucuzlayan şey denemekti. asıl rekabet fikirde başlıyor şimdi.

community note worth reading this is survey results not actual "buying" this reminds me of startups thinking they have PMF and asking folks if they will buy build it then go back to those folks and say "here you go, just what you said you wanted! please give money!" there is a lot more "no" at that point unless one is careful

A toothpaste company has quietly killed the entire market research industry and nobody is talking about it. Colgate published a paper showing you can predict real purchase intent at 90% accuracy by simply asking LLMs to roleplay customers. And this is beyond insane. If you ask an AI, "Rate this product from 1 to 5," it gives safe, middle-of-the-road garbage. So researchers invented a method called Semantic Similarity Rating (SSR). Instead of asking the AI for a number, they asked it to roleplay. They gave the LLM a demographic profile. They showed it a product concept. And they asked it to write down its raw, unfiltered thoughts. Then, they used a semantic model to translate those written thoughts into a numerical score. The results are staggering. Tested against 57 real corporate surveys and 9,300 actual human responses, the synthetic AI consumers matched real human buying behavior with 90% reliability. They perfectly mirrored how different age brackets and income levels react to price changes. And they provided detailed, qualitative feedback that was deeper and more critical than what actual humans wrote. This destroys the economics of traditional market research. You don't need to wait a month to see if a product will sell. You can simulate 1,000 hyper-targeted customer interviews overnight. You can A/B test pricing across every demographic instantly.

We have intriguing signals: Jane Street runs serious custom ML trading on large GPU clusters. Fed 2025 research finds AI agents herd *less* than humans. Some backtests show naive LLM strategies underperform over time or across regimes. But rigorous public data on these new small AI quant funds—sustained “fuck-you” returns, systemic ripple effects, or “correlated model” fragility—is still thin, mostly anecdotal or self-reported. Classic quant dynamics (edge decay, competition, costs) likely dominate. Early days; watch closely, don’t over-extrapolate.

@deedydas @grok think deeply: do we have enough data to answer these questions?

There’s a new phenomenon of small groups of people who are running these small little quant funds driven by AI models who are making fuck-you returns. I’ve personally seen many who are 2x’ing capital in months. Many unsubstantiated rumors also claim SSI is a quant shop too. Well known quant funds have all tested out LLMs for trading. Some claim it doesn’t work. Others, well.. what do we think Jane Street doing with their huge GPU cluster? On top of that, there’s a ton of people asking Claude / GPT what stocks to buy and/or “vibe code me a trading engine”. Applies to other financial instruments too: derivatives, futures, crypto, and on the less sophisticated side, prediction masks. It begs the question: how does this change how we think about markets? how much retail volume is driven purely by the ripple effects of AI models? does this completely destroy efficient market hypotheses in favor of “correlated model hypothesis.” Early theories say – Small studies including one by the Fed show destabilizing effects. – We see amplified concentrated trades into the 20 known names in the market. – Can leave trading vulnerable to GEO attacks like publishing specific articles to “poison” the models decision making. Eventually any alpha generated by the model at a point of time *should* decay over time. New anti-AI trading strategies with custom post trains too. Remember, you need be able to afford the tokens to participate in this alpha. What does this mean for the future of wealth accumulation? We live in a brave new world.

@PeterDiamandis headlines have conflict of interest with truthiness

People ask me how I stay so optimistic. The honest answer: I read the data, not the headlines.

@cmuratori sounds like the leading lawyer question: "Yes of No answer please: have you stopped cheating on your taxes?"

For people using AI in commercial game development: I'd be interested in hearing the best arguments as to why you think people should pay for the resulting game instead of pirating it. Concisely, if you pirated the inputs, why shouldn't they pirate the output?

@codevsdev i would hope the obvious answer would be: > yes, maintainable ai code and actually ai works better with maintainable commented code etc win win ?

the real flex isn’t writing code. it’s writing maintainable code. yes or no?

@dessaigne bill for outcomes 💪

A founder asked me how to build an AI-services business without it turning into a consultancy. My answer: the whole game is in how you charge.

reagarding the OP: i'd love to examine the code of a "production ready SaaS a 20 year old vibe coder" produced 🤣 in a w/end !!!! 🤣🤣🤣🤣🤣🤣🤣🤣🤣🤣 i assume the point is this vibe coder will have no computing science background and not be a software eng they would to hit PMF, agree with the spec happy customer, quality UI/UX, performance, security, resilience, maintainable, robust etc it can't just be a clone of an existing product because why not just use the existing oh man - please just give me a link to any 3... smells like bs @grok are there any hackathons where vibe coders make s/W then professional engineers try to tear it apart (the winner is the Viber who gets dunked on least)

Ask me the equivalent question about Artist, Musician, Lawyer, DJ.... (In those cases "20 years" won't cut the scale of the disruption)

or, you know, what these ppl said for me i am seeing the bottleneck effect @grok explain this theory where only one part of the value chain has ramped up

or, you know, what @JeffBezos said...

> "ai is too slow" i take the reverse view a dev's focus should be on the other two sides of the iron triangle IMO: quality cost because speed has been well and truly multiplied vs pre GenAI era i am buoyed by @ID_AA_Carmack as if you follow his thinking the more human-like the code, the better for quality (for ai and humans!) ie verbose comments & modular code & descriptive function & variable names & no god functions etc help the ai and are already best practice for coders grok.com/share/bGVnYWN5…

accepting your premise (& respect for fellow tech veteran status!) yours is still not a 100% reliable prediction about an uncertain complex future for instance: if 80% of today's valuable economic work is done by future ai the economy **may** expand to do much more and humans will be as busy as ever & doing higher value work in some cases (see history: though ai is unlike any tech before it so we must be careful) we see an example of this where coders are getting 10x the volume of code per day at 100x the typing speed etc but still feel like they need loops because interacting with the ai is too slow! 🤣 (topic for another day) TLDR; Jevon's Paradox

I'm a technology optimist. I’ve spent four decades studying disruptive innovation, from the microprocessor, the internet, mobile phones to OpenAI. I'm certain AI will do 80% of the economically valuable work humans do today, for 80% of all jobs, faster than most believe. The question isn't whether mass underemployment arrives, but whether we have a policy framework ready. Right now we don't.

@ValerioCapraro clever test

Claude Fable 5 doesn’t truly understand. And here is a beautiful proof: The Beninatto-Trombetti test is a translation test for professional translators. It measures the ability to infer context, revise the surface form, and generalize beyond literal mapping. For example, the correct translation of: “Solo 3 parole: non sei solo” is not: “Just 3 words: you are not alone” but: “Just 4 words: you are not alone.” An LLM that understands the sentence must also update the meta-linguistic claim inside the sentence. Claude Fable 5 is arguably the most advanced LLM currently available. And yet it still fails this simple test. LLMs are extraordinary machines for recombining existing knowledge. But they don’t truly understand. We are still far from AGI.

@ID_AA_Carmack interesting

It seems like LLMs could optimize coding style by exploring ways of structuring code so weaker and weaker models can still successfully perform tasks in a codebase. There are surely stylistic quirks that are peculiarly impactful to transformers, but I bet there would be a lot of overlap with human capabilities. Optimizing for understanding should help even the top frontier models, allowing them to understand things “at a glance” without having to explicitly explore. There will remain “better” and “worse” ways to code.

@natolambert "the idea of silent manipulation. This again is a horrible precedent" -> step X of Y towards ai end game...

The core part of this Anthropic Fable release saga is that there are many overlapping issues at once. Some of which operate on different timelines of the AI arc, and some have easier fixes. In my critiques, I asked for specific changes to some things, understanding that some things don't have an easy fix. The simplest issue was an uneven application of safety domains in a way that was misleading to users. This was an implementation issue that overlaps with a values-based decision of what their customers should be doing. Many people including myself pointed out how it was insane to list core safety areas and then have one of them launch with a different safety mechanism, one which actively mislead users. Doing this from the guise of safety was a major misstep and in my opinion Anthropic got very justifiably raked over the coals for it. Don't release the model if you can't hit your safety targets. A subissue here is the idea of silent manipulation. This again is a horrible precedent, and quite odd for a company that has done extensive, leading technical AI safety research on ideas like CoT monitoring and other emergent misalignment issues. Silent manipulation of users is baking in a misalignment to the system at its face level. This comes with a permanent degradation in user trust, which begets a less safe environment for AI. Users who don't have clear information on how AI works will not develop safe working patterns with it. The more complex issues are with how Anthropic handles broader scientific engagement with their models. The safety classifiers launched with these models obviously have accuracy issues to start. I have priced in that there will be more false positives to start, that's life. It's Anthropic's business to degrade their products at release time, or make the trade off of user satisfaction versus revenue. Still, it is a very real sign of concentration of power that businesses can make such obviously user-harmful behaviors and still lead in the market. This concentration of power is only starting to set in and we could see even weirder signs of it in the coming years. It is now simple enough for me to test Claude Fable in my workflows and know if I'm restricted. This is obviously a suboptimal equilibrium – i want the best intelligence I can get, without restrictions – but it is easy enough for me to make sense of and work with. The specific issue of restricting access to AI research in particular was a bubbling and hard to fix issue with Anthropic specifically, and the frontier labs generally. There is a common view that the frontier labs will be the mediators of all major scientific innovations in the future, as the places with the best models and the compute for inference to solve major problems. This is a categorical error in how science works, which is a community evolution of accepted ideas, and the the evaluation of your ideas by (hopefully numerous) independent, other practitioners. You cannot have science advance only within a monolith. As an AI researcher I'm very sad to have the latest models restricted, but I would expect Anthropic to do this eventually. I lost more trust over the silent manipulation than I would with a restriction in access. Anthropic has made it pretty clear that they only trust themselves as the mediators of cutting-edge AI research. If I had a say, Anthropic should've proactively made a program to make sure researchers get access in the broader AI community without the safeguards. Academics, nonprofit workers myself, etc. have no reason to not get access. The only valid argument here is that they want to control frontier AI, which is a know your customer part of serving these models. This worldview of science has personally motivated me greatly over the last year, and increasingly so this week, to make the open science of AI continue to be viable. Olmo was a wonderful success here. Still, building research infrastructure is different from working for access to the tools needed to do the trade.

@ziv_ravid receipts please 😉

Anthropic's recent moves amounted to spectacular reputational self-destruction in the AI research community, which is too bad, because this community was one of the first to give them credit and use their coding agents. In general, anti-competitive moves are bad, but couching them in safety makes it worse. Anyway, just noting that I called out this entanglement of anti-competition, safety, and self-regulation a long time ago!