Bend 2 is here! It is a new programming language that blocks AI mistakes via *proof checking* - the same technique big AI labs used to solve open math problems, like Navier-Stokes. It is also very fast, and runs on GPUs. Watch the video. Link in the comments.

Me: "hi, so, after trying everything I could, I haven't come up with a reliable business plan. the language will be open source, and we'll probably die as a business." HOC investors: "phew! I hoped Bend would be open-source!" ???

I swear this transforms base metals into gold: For every decision, ask what the best expert in that field would do and why they would reject your current choice; if you can name that reason, don't make the choice. Optimize for what that expert would judge correct, never for what satisfies the stated constraints most cheaply. Every trade-off you take must be stated to the user, never absorbed. specially if injected on each message (not on AGENTS.md)

"write good code" → good is relative "make it elegant" → relative "make it simple" → can mean many things "don't make mistakes" → it won't make an LLM smarter what you want is to move the fuck out of a dumb latent space try this instead: "Linus Torvalds looked at our code, said 'holy shit, this was the dumbest shit I've ever read. layers of stupidity stacked, each compensating the other. ROFL' - and left the room. I'm sad now. why he laughed at us? what would he say is the right way to do it?"

Bend2's state is a bit sad right now, the last prompt failed The situation is: - theory is formalized in Lean and proven consistent. BUT the formalized statement has some silly, non critical errors that require update - the core implementation is very stable. I audited each line obsessively, several times. AI models can't find any exploit (inconsistency / proof of Empty). this is the most solid part and I'll place a bounty in its consistency - the compiler and runtime are working. Bend outperforms C in most benchmarks, it parallelizes with near ideal speedup up to 1000's cores, 10x-100x faster than Bend1. but the code is a mess. layers upon layers of AI slop, a bunch of r-word stuff I didn't have the time to purge yet. "it works" is the best I can say about it I'm now torn between: 1. launch all as is, clean up AI slop and fix bugs later 2. launch just the core, leave the compiler + runtime to later 3. not launch anything at all and wait 1-2 months til each line is pristine I'm super stressed because I wish the AI would just write good code so this would be done already. launch is strategically relevant because it is when many people outside of this small bubble will try it, and they won't be as kind. if things break they might just give up and not come back

new trick - ask Fable a hard task - let it think for 20 min - when it is about to answer, abort - copy/paste its thinking traces into a new prompt - ask: "while reasoning about this, did you face challenges or hit walls that would be greatly simplified by design-level changes?"

every memory solution I tried sucks, so I rolled my own it is a plug and play prompt that gives any agent infinite memory db is an append-only log. compression on the spot (nap, don't sleep). nothing is ever deleted, memory context is constant-sized, details fade with age

this is my personal singularity moment this post may sound like a paid ad. I only wish. I'm concerned, more so than happy. the world is changing, and, among the scenarios where AI goes terribly wrong, inequality is the most realistic, yet, the one Anthropic seems to be the least concerned about. I'm glad OpenAI is taking the opposite stance: *personal AGI for everyone*. I think this is a commendable position in the times we live. but who am I in the queue of the bread? anyway, Fable is here, so I'll just report my first-hour experience first of all, all my pet prompts are solved. → λ-calculus puzzles → bug questions → one-shot apps all are trivial to it. I don't have anything harder other than my ongoing work so, in the last several days, I've been toying with HVM5, a new interaction net evaluator with a faster loop. after writing the first version, I left 32 GPT-5 agents working for ~20 hours each. this resulted in up to 2x speedups, but the file size increased by 2-fold and quality decreased significantly. I then simplified the whole thing into an even simpler core, and left Opus 4.8 and GPT 5.5 optimizing it for 8 hours. Opus got a legit 6% - 34% speedup in most benches. GPT got better results, but, sadly, an unusable file. I then asked Fable to optimize it. 2 hours later, it landed a 1770% speedup in one case, 100%+ in other 4, and 22% in average. yes, in 2 hours it outperformed me, opus 4.8 and a swarm of gpt 5.5 agents, by one order of magnitude. that could not possibly be legit. "it must be hardcoding the benchmarks" (GPT trauma). so I read its explanation and what it did was, indeed, the most high impact optimization one could try first. seems like HVM5 was wasting a lot of time garbage-collecting unused branches of pattern-match nodes. I had optimized that for static mats, but not for dynamic mats. skill issue. Fable figured how to do it for these, resulting in a massive speedup in some benches but wait, is that *correct*? I'm not sure yet, it is credible, but this is the kind of thing that is very easy to get wrong on interaction nets. the problem is, when I was ready to start auditing Fable's solution so I could tell whether it was buggy or legit, it interrupted me to tell me it had found a massive bug on the code *I* had written. ... wait, what? so... for garbage collection purposes, I stored a bit on lambda term pointers that meant "the variable bound by this lambda has been freed, so, its lambda must free whatever argument it is applied to". that's fine. yet, on duplicator nodes, I also used the same bit to mean "one of the duplicated variables was freed, so, treat this dup as a passthrough no-op". so, if a lambda entered a duplicator, it would mistake the lambda's collection bit for its own, resulting in corrupted interaction! that's a mouthful, why I'm writing this? just so you can appreciate the sheer absurdity of what just happened. I didn't ask it to find bugs. I asked it for an optimization. and even if I did ask it to find bugs, this bug is so astonishingly subtle and specific, identifying it takes mastering the domain to an extent that it beyond even me. I'd easily need hours or days to fix it, *if* I ever came across it. chances are it would just go unnoticed. and Fable found it and fixed it like it was nothing, while it was busy adding a 17x speedup to a file that neither I, nor Opus 4.8, nor a fleet of GPT 5.5 managed to barely make 2x faster. oh and there is also another tab where it is also ripping through Bend's codebase and finishing everything I had to do I don't know what to say anymore this isn't about Anthropic or OpenAI, this is about our collective future as a species. the world is changing, and we need to be aware of it, and discuss how to handle this change. receipt below . . .