It's not clear to me how or why this works, and how it compares to just using md files in my project. For something like this, we really need benchmarks.
Hey man, if it works it works. There's a reason everyone is creating AI tools. We're all buying them. I'm still waiting for someone to make a world-class cli harness that can replace Claude Code but solves the memory and design problem. Web design is still a nightmare with LLMs.
They don't make sense, they're a lie that these AI companies keep spamming using bots so that useful idiots perpetuate it, so that they can keep draining us of money. Straight out of the Anthropic handbook. They've always been cheap to run. I wouldn't be surprised if Anthropic is running for <$1 for 1M/tok.
No, but the AI labs would love to frame it this way so they can continue to nerf models and increase prices while they use the cheap, highly performant, highly powerful models internally to replace all of your businesses.
The Twitter example is a bad one. Elon Musk, at the time, was making hundreds-of-millions through crypto-market manipulation on Twitter. At that time he realized that having control over the entire Twitter platform would unlock many billions of dollars worth of profit opportunity. Attention is the most valuable and powerful currency in this world. Not only for manipulating markets, but also for political propaganda. The information we consume literally shapes the world. So yes, it was a 4D chess move.
Has anyone ever done a proper security audit of VLC that is downloaded from the web? I don't trust it, and the fact that their releases on Github don't include binaries makes me trust it even less. Nobody is compiling VLC from source, and they don't provide any sort of provenance from the GH actions pipeline.
Thinking time is not the issue. The issue is that Claude does not actually complete tasks. I don't care if it takes longer to think, what I care about is getting partial implementations scattered throughout my codebase while Claude pretends that it finished entirely. You REALLY need to fix this, it's atrocious.
Do you guys realize that everyone is switching to Codex because Claude Code is practically unusable now, even on a Max subscription? You ask it to do tasks, and it does 1/10th of them. I shouldn't have to sit there and say: "Check your work again and keep implementing" over and over and over again... Such a garbage experience.
Does Anthropic actually care? Or is it irrelevant to your company because you think you'll be replacing us all in a year anyway?
Whenever Anthropic has an opportunity to do what's right, they go the opposite way. For example, their source leaks, and instead of open-sourcing it like people have been asking to happen for years so they can contribute fixes because Anthropic doesn't care to maintain their own software, they tighten the noose further.
If it isn't obvious by now, this problem is only going to get worse. The only reason we have subscriptions still is because they're waiting to pull off the biggest bait and switch in history. Don't get sunk on this ecosystem, or you're in for a world of pain in the future. As has always been the case; competition and open-source are our only hope.
They're just removing it from public access and selling it to big money instead. Think large advertising companies, government agencies, Coke-Cola, Hollywood, etc. The scary part is now that they've removed it publicly, it's going to be harder to keep a pulse on what is real and what is fake. We can't trust any video, audio or text content now.
Does it matter though? If it accomplishes the task, it accomplishes the task. Everyone uses a harness anyway, and finding the best harness is relevant. Also perhaps this hints at something bigger, i.e.: we're wasting our time focusing on the model when we could be focusing on the harness.