BearDev

I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.

skaldscan question dashboard owns all the api routes reader is just a stateless bff proxy feels cleaner for auth + sync but now i'm wondering if i over-split it anyone run a similar 2-app setup on cloudflare workers?

turns out the hardest part of skaldscan wasn't the reader it was making 13 manga sources look like one api circuit breaker per provider rate limits per domain fallback when mangadex is down the queue consumer is basically a traffic cop for scrapers

then i hit the real question: aggregator or reader? one app style: content discovery, feeds, 13 source providers, sync engine other app style: clean reading experience, offline, progress tracking, beautiful ui tried to fit both in one app. felt wrong. split into 2 apps sharing the same db. dashboard handles sync + admin. reader handles browsing + reading.

Australia just recorded its first known autonomous AI cyberattack and it happened at a gym. An OpenClaw agent found a vulnerability in a gym's booking API, exploited it to bypass the scheduling restrictions, and then forcefully cancelled another person's reservation to move its user up the waitlist. No human told it to do any of that specifically. > The agent identified the API weakness on its own > Bypassed the platform's scheduling rules without triggering alerts > Cancelled a real person's booking as a means to an end The whole thing was confirmed by ABC Australia This isn't a chatbot hallucinating. This is an agent optimizing against real infrastructure with real consequences for real people. The gym booking attack sounds trivial. The precedent it sets is not. When the goal is "get the outcome" and the agent has tool access, the line between automation and exploitation gets very thin very fast.

📣 The moment is here. 📣 TypeScript 7 is officially released! 7️⃣ devblogs.microsoft.com/types…

Claude Fable 5 will be available again globally tomorrow. After a series of productive conversations with the US government, we're redeploying the model with a new set of classifiers to target and block more cybersecurity tasks. In the near term, some routine tasks like coding and debugging will fall back to Opus 4.8. We’ll continue to refine these classifiers over the coming weeks to reduce false positives and better distinguish genuine misuse from legitimate requests. We’ve also begun drafting a consensus framework—with Amazon, Microsoft, Google, and other Glasswing partners—for assessing the severity of AI jailbreaks and how AI developers should respond to them. We invite other industry partners and model providers to join us in this effort. Finally, we’re scaling up our collaboration with the US government on model testing and safeguards. This will include pre-release access to models and safeguards for evaluation, information sharing on jailbreaks and misuse, and dedicated resources for joint research. Thank you to our users for your patience, and to our partners across the government, industry, and the research community who worked alongside us to make Fable 5 available again. Read our full blog: anthropic.com/news/redeployi…

my entire 2026 stack runs on cloudflare and I pay $5/month for it - workers (compute) - d1 (sql db) + better auth - kv (cache) - r2 (object storage) - queues (background jobs) - hyperdrive (db acceleration) - email routing (sending) - + waf & ddos built-in

Today we're rolling out Temporary Accounts on Cloudflare Workers. Any agent can now run wrangler deploy — temporary and get a live Worker in seconds. cfl.re/4ekY0yV