TLDR AI 2026-08-04

PREPARE FOR THE ERA OF AI-DRIVEN EXPLOITS.

FORMER OPENAI EXEC FIDJI SIMO DISCUSSES HER BATTLE WITH POTS AND HER STARTUP'S PLANS TO CURE IT WITH AI AND 3,500 VIALS OF BLOOD

Fidji Simo left OpenAI's leadership team after a seven-year battle with Postural Orthostatic Tachycardia Syndrome (POTS). Her new startup, ChronicleBio, will focus on using AI to cure POTS and other chronic diseases. The company has collected 153 terabytes of data from blood draws from people with chronic diseases so far. It plans to launch home blood draws to further increase its dataset. ChronicleBio plans to use the data to learn more about diseases and improve the success of clinical drug trials.

ANTHROPIC PAYS AI'S BIGGEST SALARIES. ITS CEO JUST DISCOVERED PEOPLE MIGHT TAKE THEM FOR THE MONEY

Anthropic's CEO, Dario Amodei, recently said he was worried that new hires were joining his company for the money rather than the mission. Anthropic reportedly pays more than any lab in AI. Hiring top researchers is hard, and keeping them is even more difficult. When every lab can pay millions, mission is the only lever left. Researchers chase money, but they also chase compute, influence over what gets built, and the freedom to work their own way.

MIND LAB PUTS CONTINUAL LEARNING TO THE TEST WITH MACARON-V1

Mind Lab claims its Macaron-V1 model surpasses GLM-5.2 in its benchmarks. The model was built by attaching five LoRA expert modules, each with about one billion parameters, to GLM-5.1. The system dynamically switches to the expert model best suited to the task it is given. The accumulated data from model use can be distilled into a dedicated LoRA adapted that is continually updated as the model is called.

HOW OPENAI BUILT GPT-LIVE

OpenAI rebuilt its voice architecture around a full-duplex model that listened and spoke simultaneously. The system combined stateful inference, asynchronous delegation, dynamic context management, and low-latency media transport to keep conversations responsive while supporting advanced reasoning and tool use.

ONE AGENT, EVERY SURFACE: HOW WE BUILT THE KIRO AGENT HARNESS

Kiro is an agentic IDE with features such as specs, steering, and hooks. The Kiro agent harness is a lightweight server-side process that runs alongside codebases, starts quickly, and owns everything on the agent side. The IDE, CLI, and Web clients own how the user interacts with the agent and how it presents the agent's work. The only way to cross that boundary is through the defined protocol interface. The well-defined interface between server and client means the agent code evolves independently of the clients.

OPENAI'S UNRELEASED MODEL ASTRA SOLVES TEN MAJOR OPEN MATHEMATICS PROBLEMS

OpenAI's recently released solutions for ten major open mathematics problems show that AI is now superhumanly capable at cyber and coding and superhuman at advanced math, the same way non-AI computers have been superhuman at basic math for a long time. These problems were well-defined, formalized problems where the solution could be easily verified. A lot of what constitutes AI R&D is verifiable. The lab that gets traction on true AI R&D self-improvement loops will find themselves in an overwhelmingly strong position.

ORCHARD

Orchard is an open-source agentic modeling framework. Its foundation is a thin, Kubernetes-native environment service that exposes generic primitives with no assumptions about the harness, trainer, inference backend, or task domain sitting above it. This foundation allows for recipes that use the same substrate for trajectory distillation, on-policy RL rollouts, and evaluations. Datasets, training recipes, and evaluation protocols stay portable across harnesses, domains, and projects instead of being rebuilt for each new study.

MIRRORCODE

MirrorCode is a benchmark that tests AI models on long-horizon tasks. It contains tasks where AI models have to reimplement entire programs end-to-end without access to the original source code. AI-generated solutions must match the original program's output exactly on end-to-end tests. The benchmark's 25 target programs span different areas of computing, including Unix utilities, data serialization and query tools, bioinformatics, interpreters, static analysis, cryptography, and compression.

FROM RLVR TO RLSVR

RLSVR expanded RLVR beyond inherently verifiable problems by transforming open-ended tasks into proxy environments with rules and outcomes that generated their own reward signals. SpyRL demonstrated the approach through multi-agent self-play, where predetermined roles and voting made evaluation automatic.

FAST GEMMA'S VERIFIED INFERENCE OPTIMIZATION RECIPE

VIDRAFT documented the full configuration behind its verified state-of-the-art Fast Gemma submission, explaining how each software optimization increased tokens per second on Gemma 4 E4B running on a single NVIDIA A10G.

GPT-5.6 SOL USES TWICE THE TOKENS OF GPT-5.5

GPT-5.6 Sol xhigh now uses more than twice as many tokens per session as GPT-5.5 xhigh in Codex workflows. This cuts the effective value of a token-based quota by more than half. At the same token price, 2.25x the tokens means roughly 2.25x the cost for a similar token mix. GPT-5.6 Sol also adds a cache-write charge that GPT-5.5 didn't have.

WHITE HOUSE TO HOST AI COMPANIES TUESDAY TO REVIEW NEW MODEL-TESTING FRAMEWORK

The White House will meet with AI companies to discuss a new framework for reviewing cybersecurity in AI models. This voluntary framework, ordered by President Trump, allows AI developers to share models with the government to evaluate cybersecurity risks. Companies like OpenAI, Google, and Anthropic are set to attend, with the assessment criteria remaining classified.

engineers work today. Rebuild your 1st interview - free

GOOGLE WORKSPACE PLUGINS

Cursor can now read, write, and act across Google Workspace through new plugins available on the Cursor Marketplace.

GOOGLE IS WORKING ON PLUGINS FOR GEMINI ENTERPRISE

Plugins may act as mini-apps or packaged workflows.

DEEPSEEK'S NEW AI MODEL IS BY FAR THE CHEAPEST OF WELL-KNOWN MODELS TO RUN, RESEARCH FIRM SAYS

DeepSeek's V4-Flash AI model is the cheapest to run among well-known models, costing 105 times less than Anthropic's Claude Fable 5.

Stop babysitting your LLM providers.

CHINA'S MINIMAX H3 IS THE FIRST OPEN MODEL TO TOP AN AI VIDEO RANKING

Artificial Analysis ranks MiniMax H3 first in Video Editing, second in Text-to-Video, and third in Image-to-Video.

BRINGING HUMAN TASTE TO AI MODELS

Design Arena raised $7.9 million to help AI companies measure subjective qualities such as whether a generated game feels fun.