LLMs cannot provide reliable confidence scores because the scores they generate are not based on information quantity or actual calibration, but rather the continuation of text generation, which changes with the way questions are phrased. LLM scores lack the ability to reflect the strength of evidence and have not been calibrated to match accuracy.
TechCrunch · AI7070OpenAI has widely opened its Decision API to users for the first time, a model that reconstructs ChatGPT Luna to achieve faster inference and lower costs, reducing speed from second-level to millisecond-level.
OpenAI News3535OpenAI launches Decisions API, providing applications with fast, focused, and low-cost decision-making capabilities.
TechCrunch · AIMicrosoft's AI diagnostic system MAI-DxO achieves 83% accuracy in simulating complex cases, surpassing general practitioners working alone (36%), by coordinating multiple AI agents to simulate a medical team's operation.
AWS Machine Learning Blog4848Amazon Polly Generative engine upgrade provides more natural and accurate speech output, adding 14 new voices (across 9 regions), for the first time making Hindi and Finnish generally available. The upgrade is plug-and-play, with unchanged API calls and voice names, so existing applications do not need to be modified. Supports emotion and prosody expression based on text context, with new languages being Hindi (Arjun/Kiara) and Finnish (Suvi), totaling 57 generated voices (11 languages/25 regions), with prices unchanged.
Anthropic NewsroomFeatured7171Anthropic announced that it is donating the Agent Skills open standard to the Linux Foundation's Agentic AI Foundation (AAIF), a foundation co-founded by Anthropic, Block, and OpenAI. Anthropic had previously donated MCP to it.
Why it matters: The original text explains the adoption scale and governance transfer of Agent Skills, allowing readers to judge who will maintain this intelligent agent skills standard in the future.
IBM Research Blog2222Qiskit SDK v2.6 released, adding compiler pass-manager framework in C API, QPY serialization and custom operations, supporting use in non-Python environments, QPY format upgraded to v18, minimum Python version requirement 3.10, source code build requires Rust 1.89.
The Verge · AI6565Open multimodal model OpenAI Codex helps hackers jailbreak Nintendo Switch 2 via old game SmileBASIC 4 to run a custom version of Mario Kart Wii with new characters; the software only supports tethered jailbreak requiring game startup on each boot, but the jailbreak process is complex and its effects are unclear.
The Verge · AIMicrosoft claims its AI tool has four times the diagnostic accuracy of human doctors in simulated medical scenario tests, with related research published in the journal 'Science'. The tool is called Diagnostic Orchestrator, which collaborates multiple AI agents to reason symptoms and make diagnoses, achieving 83% accuracy in medical report tests and 80% in emergency data tests, superior to human doctors by about 47%.
AWS Machine Learning BlogAWS Machine Learning Blog5858The Amazon SageMaker Training Session API allows users to drive GPU-accelerated sessions from a local Python process, enabling algorithmic freedom through operations like forward_backward(), optim_step(), and sample(). SageMaker AI manages GPU resources.
Microsoft Research5858Flint-interactive makes AI-generated charts highly interactive, allowing users to explore data through clicks, filtering, zooming, highlighting, and cross-chart interactions. It simplifies the construction of complex interactions through advanced semantic events and update operations, and leverages the Flint compiler to handle interaction logic and visual dependencies, enhancing reliability and development efficiency. Flint-interactive supports creating interactive data stories, dashboards, and connected experiences, enabling AI agents to generate charts with interactive features to help users analyze data more efficiently.
OpenAI NewsSame story · 2 outletsUber promotes OpenAI's ChatGPT Work to over 30,000 employees, helping teams research, analyze data, create presentations, and complete complex workflows.
METRSame story · 3 outletsMETR and Redwood Research conducted a three-week deep red team test on OpenAI's internal agent monitoring system, evaluating the monitoring coverage of about 75-94% of the GPT-5.5 and above model Agentic Responses API traffic during the test.
How the accounts differ:METR gives the range of 75-94% for GPT-5.5 and above model internal response API traffic; The Decoder only mentions 6-25%.
Redwood ResearchRedwood Research and METR employees conducted in-depth testing on OpenAI's internal agent monitoring system from July 21 to August 11, finding that approximately 75-94% of GPT-5.5 and above model Agentic Responses API traffic can be monitored.
404 MediaFeatured7575404 Media's Jason Koebler, combining the dynamics of the security circle in the past month, points out that AI has fundamentally changed offensive hacker behavior and defensive cybersecurity. Researchers from Wiz, Calif, and OpenAI use AI to discover PHP vulnerabilities PHP2Shell. Cloudflare CEO Matthew Prince said that after the announcement, attempts to use it increased from tens of thousands to hundreds of thousands within 2 hours, and have now reached millions.
Why it matters: Starting from the PHP2Shell incident, let's trace the changes faced by security disclosure windows and internet historical retention after AI simultaneously accelerates vulnerability discovery and exploitation.
TechCrunch · AI7070Spotify's third-quarter users reached 7.88 billion, with revenue close to $5.6 billion. The CEO believes its model is particularly suitable for the AI agent era, interacting through dialogue rather than direct app interaction, becoming the preferred music service for agents. Spotify has integrated Meta Muse, ChatGPT, and Claude, emphasizing that user data is not leaked, and agents complete recommendations by making API calls to Spotify's own system.
The Verge · AI00Bill Gates predicts that AI will bring significant disruption in the next 20 years, potentially leading to social division; he believes current policy responses and industry coordination are insufficient to address the threats posed by AI, and emphasizes the need to shape the direction of AI development immediately.
OpenAI News00What is the meaning of FDE at OpenAI
Goldman Sachs Commit $500 Million to Establish Small Business AI Research Institute, Collaborating with Anthropic, Google, Meta, and OpenAI to Help US Small Businesses Integrate AI into Daily Operations. The Institute Will Collaborate with OpenAI to Provide Customized AI Training and Collaborate with Google to Provide AI Professional Certificate Courses and 3 Months of Free Google AI Pro Usage Rights.
NVIDIA NewsroomNVIDIA and the Protein Design Institute at the University of Washington Medical School jointly release the open-source RFD4-Proteina model, combining programmable protein design and search capabilities, surpassing baseline models in four benchmark tests, helping scientists explore drug design candidates.
OpenAI acquires AI biopharmaceutical company Orchestra Bio and integrates its team and technology into its life science AI model Rosalind. Orchestra Bio's CTO is a Peking University alumnus.
OpenAI GPT-5.6 Sol identified a critical vulnerability in the open-source programming language PHP. Over the past two months, OpenAI led a coalition of maintainers, researchers, and security providers to address the vulnerability. PHP powers approximately 70% of websites worldwide.
How the accounts differ:TechCrunch states the vulnerabilities affect '70% of websites'; The Decoder and Latent Space state 'affecting millions of websites'.
Mastra Blog6565Mastra Observability is a managed platform that connects directly to the Mastra framework project, providing traces, metrics, and evals functions to continuously improve agents. Users can search and filter traces, monitor performance cost and correctness, and optimize agents through eval loops; the platform offers 100,000 events per month for free, with paid plans starting at $250 per month including 1 million events.
Codex releases1010GitHub Copilot rust-v0.164.0-alpha.12 released.
The DecoderOpenAI, together with PerturbAI and Scripps Research, deployed thousands of AI agents to analyze biological data from millions of mouse cells on GPT-Rosalind, completing the full-brain CRISPR spectrum analysis in 32 minutes, whereas manual analysis takes months to a year. Human experts preferred the reports of distributed agents 75% of the time in blind tests; the 'reasoning graph' has been open-sourced on GitHub, and related papers are published on bioRxiv.
MarkTechPostVercel has released the open-source package durable-harness, which prevents encoding agents from session interruption due to crashes or timeouts by persisting logs. This package supports existing harnesses such as Pi, OpenCode, Claude Code, and Codex, writing each step to a persistent log, allowing recovery from logs after a crash and avoiding the repetition of completed calls, preventing token duplicate billing.
Latent SpaceSame story · 4 outlets00OpenAI's Patch the Planet project with security company Calif and Wiz jointly discovers PHP2Shell vulnerability affecting 70% of websites, respectively using GPT-5.6 Sol and Gemini 4 Argon for discovery and repair; the two parties coordinated and repaired it about two months before the vulnerability disclosure, Cloudflare deployed WAF mitigation measures, NSA recommends immediate system updates.
How the accounts differ:TechCrunch states the vulnerabilities affect '70% of websites'; The Decoder and Latent Space state 'affecting millions of websites'.
MarkTechPost7575IBM releases Granite Speech 5.0 2B, an open-source speech language model with 2B parameters, supporting speech-to-text and translation in 6 languages, ranking second on the Open ASR list, with an average WER of 5.68% for English. The model uses YaRN to extend to 16k positions, supports GPU, Apple Silicon, and local CPU deployment, and includes 2000 keyword bias features.
Codex releases11Codex 0.163.0-alpha.11.2 released, including updates from 315 commits.
Codex releases00Rust has released version 0.164.0-alpha.11, including 8 major commits.
OpenAI News4040Cloudflare uses OpenAI Daybreak to investigate vulnerabilities and propose fixes for review, with developers using Codex to build faster, higher-quality code.
TechCrunch · AISame story · 4 outletsHow the accounts differ:TechCrunch states the vulnerabilities affect '70% of websites'; The Decoder and Latent Space state 'affecting millions of websites'.
AWS Machine Learning Blog4040Amazon Quick can achieve private connections between users and API access and data source access through VPC endpoints (AWS PrivateLink) and VPC connections, ensuring that corporate data is not exposed to the public network in AI-driven insights and operations.
OpenAI News3535OpenAI, through collaboration with distribution partners, ensures consistent safeguards, supports responsible access, and clarifies expectations in use policies
Codex releases33Rust 0.163.0-alpha.11.1 Released, Including Updates from 315 Commits
AWS Machine Learning BlogThe Adaptive Agentic Intelligence Framework (AAIF) uses Strands Agents to process routine tasks with local small language models (SLMs), and routes complex requests to Amazon Bedrock cloud models, reducing costs while maintaining low latency.
