The Brief
New briefs daily around 7 AM Eastern
Full story
Together AI says agent-leaked key exposed June and July prompt samples
The attacker downloaded stored prompts from MiniMax M3, GLM-5.2 and Kimi K2.6 requests, then tested credentials found inside them. A secret pasted into one of those prompts became a working key for someone else, so the leak reaches past the prompts into customer systems.
- 01Read
MIT CSAIL: REPL harnesses let models solve tasks 64x longer than trained
A 27B model trained with synthetic surrogate harness tasks raised its long-context average 26.9% on about half the tokens. Context kept in a REPL makes each model call look like a short familiar task, so cheap short-task RL carries over to much longer work.
- 02Read
Frontier models rewrite a 0.5B vision model's weights with no training
The calculated edits reached 98% accuracy on mockingbird vs noddy and raised camera-trap scores in about an hour. That puts direct edits beside LoRA fine-tuning as a way to adapt a small model, and the released code and comparison let others test the claim.
- 03Read
Tool-free LLM reads, patches and writes executables from hex dumps
The setup was audited to strip tool access, so the results come from the model reading bytes alone. Malware triage and binary patching normally start in a disassembler, and here the model explained five binaries, found indicators and added features without one.
- 04Read
Figma runs its AI-code sandboxes through one platform called Nimbus
Figma Make and the design agent execute untrusted generated code in tens of thousands of sandboxes with differing memory and network permissions. One API now creates, reuses, monitors and deletes them across several cloud providers, keeping those products available.
- 05Read
Willison: Gemini 4 Hydrogen at xhigh reasoning costs about 6x high
In his comparison xhigh used 31,477 output tokens for 23.61 cents against 5,311 tokens and 3.99 cents at high. llm-gemini 0.38.1 now passes xhigh to Hydrogen and Neon while Carbon rejects it, so the top setting is a sixfold cost step.
- 06Full story
PHP fixed a critical session flaw. What to ask whoever runs your site
An interrupted session write can splice two valid sessions into attacker-controlled data that PHP then deserializes, a path to remote code execution. Cloudflare's WAF rules reduce exposure, but only releases 8.2.36, 8.3.37, 8.4.28, 8.5.13 and 8.6.0 fix it.
A dated brief from the vibe-coding frontier. Today’s Intel.