Live intelligence

Latest AI updates

186 updates

Products Editor selected

Together AI Releases Together Compiler: Agents Make Inference Up to 2.6x Faster

Together AI introduced Together Compiler (TKCC), an agent-driven system that continuously re-optimizes inference as production traffic changes. It uses coding agents to rewrite a model's forward pass)Skip and a persistent engine with hot reload to measure candidates, achieving up to 2.6x higher throughput than best published configurations within 24 hours. The system runs on multiple hardware platforms and has been tested on several models, with gains ranging from 4% to 2.6x depending on the setting.

Why it matters: This is a concrete, recent AI development with measurable performance gains and a novel approach to inference optimization, offering practical value for developers and researchers.

MarkTechPost7h agoScore 85

AI inferencecompilationagent-based optimization

Products Editor selected

OpenAI Launches GPT-Image-2.6 Flare and Sunburst with GIF, SVG, and 6-Second Generation

OpenAI released GPT-Image-2.6, a new image generation model family available as two API models: Flare for speed and Sunburst for detail. The update introduces GIF output, SVG export, markup-guided edits, and a complexity parameter, with Flare capable of generating images in as little as 6 seconds at low quality. OpenAI claims 50% faster generation than the previous version, and usage has grown to over 4 billion images per week. Pricing remains token-based, matching GPT-Image-2.5 rates.

Why it matters: This is a concrete product launch with new capabilities (GIF, SVG, complexity control) and performance claims, offering practical value for developers and AI enthusiasts.

MarkTechPost8h agoScore 82

OpenAIimage generationAPI models

Products

AgentTrustCI: Build-Time Policy, Grounding, and Trajectory Tests for Java LLM Agents

AgentTrustCI is a Java/JUnit 5 framework that adds application-layer behavioral checks for LLM agents, covering tool selection, answer grounding, and workflow trajectories via recorded traces. It uses deterministic replay for repeatable build-time tests and an optional judge interface for semantic evaluation. In a synthetic corpus of 55 cases across seven enterprise domains, the full pipeline correctly classified all 30 injected faults and 25 passing controls, including 10 non-trivial controls, with no missed critical cases. The paper emphasizes that results reflect the authored regression suite, not production accuracy, and describes a production-trace validation protocol.

arxiv10h agoScore 78

LLM agent testingbuild-time validationbehavioral contracts

Products Editor selected

GPT-Image-2.6全量上线：直出GIF和SVG，6.2 Astra同步现身

OpenAI发布GPT-Image-2.6，支持直接生成GIF和SVG格式，最快6秒出图，并同步推出6.2 Astra模型。该版本在图像生成速度和格式支持上实现突破，为创意工具和开发工作流带来新可能。

Why it matters: 该更新涉及具体模型版本和功能增强，对AI图像生成领域有直接实用价值。

qbitai13h agoScore 85

OpenAI图像生成模型更新

Products Editor selected

AT&T's Connected Life integrates Gemini for Home

AT&T's smart home security system, Connected Life, now integrates with Gemini for Home, claiming to be the first home security product to do so. Users can get AI-powered summaries of home events, search video footage using natural language queries, and receive detailed descriptions of security camera events. The integration leverages Google Home Premium to bring Gemini features to AT&T's security ecosystem.

Why it matters: Shows a concrete, consumer-facing AI integration into home security, demonstrating practical value of Gemini for everyday users.

The Verge AI18h agoScore 85

AT&TGeminismart home security

Products

ChatGPT adds animated GIF creation with OpenAI's Images 2.6 model

OpenAI has added a GIF feature to ChatGPT powered by its Images 2.6 model. Users can generate animated GIFs from text prompts or photos, then adjust playback speed and individual frames. The model's image consistency keeps characters recognizable across frames, enabling flipbook-style animation. OpenAI says users now generate more than four billion images per week.

the-decoder19h agoScore 74

OpenAIChatGPTimage generation

Products Editor selected

Anthropic's text watermark is easy to bypass on several older Claude models

Anthropic is rolling out text watermarks to four more Claude models ahead of the EU AI Act's December deadline. However, on older models like Claude Opus 4.6 and Sonnet 4.6, API users can disable the watermark by changing a single sampling parameter. Anthropic's response to the loophole is a compliance warning rather than a technical fix.

Why it matters: Highlights a practical loophole in AI watermarking compliance that affects real-world deployment ahead of regulatory deadlines.

the-decoder20h agoScore 85

AnthropicClaude modelsAI watermarking

Products

Nothing teases a screen-free AI device ahead of launch next month

Nothing CEO Carl Pei announced on X that the company's first AI-native product will ship in December, over a year after revealing plans for AI-native devices and an AI OS. The device will have no screen and is described as a tool with Nothing's signature personality, not an app. Pei shared a teaser image with the 'NOTHING(R)' branding, but few details are known.

The Verge AI20h agoScore 78

AI hardwareconsumer devicesproduct launches

Products Editor selected

ChatGPT will now make you animated GIFs

OpenAI has rolled out GIF generation in ChatGPT, powered by its new Images 2.6 model. Users can create animated GIFs from text prompts or uploaded photos, adjust playback speed, edit individual frames, and share the result. This marks the first time ChatGPT can natively produce GIFs in a single step.

Why it matters: Provides a concrete, useful product update that expands ChatGPT's creative capabilities for everyday users.

The Verge AI21h agoScore 85

OpenAIChatGPTimage generation

Products

I tried Apple’s Siri Recaps. They’re surprisingly useful — and a little unsettling

The article is a hands-on review of Apple's Siri Recaps, an AI feature that summarizes notifications and messages. The author found it surprisingly useful within a day, despite initial privacy concerns, but notes there is room for improvement. The piece highlights both the practical benefits and the unsettling aspects of AI-driven personal summaries.

TechCrunch AI22h agoScore 78

Apple AISirinotification summaries

Products Editor selected

Smart Approvals in Amazon Quick: Human oversight where it matters

AWS introduces Smart Approvals in Amazon Quick, a feature that uses a classifier model to automatically approve eligible agent actions or return them for human review, balancing oversight with reduced interruptions. It evaluates actions against user requests and conversation context, respects existing permissions and mandatory approvals, and provides audit trails via CloudTrail and CloudWatch Logs. The feature is on by default, configurable via Custom Permissions, and designed to mitigate prompt injection and preserve human control over consequential decisions.

Why it matters: This is a concrete new AI governance feature from AWS that addresses a real challenge in AI agent deployment, with clear technical details and practical guidance for security leaders.

AWS Machine Learning Blog22h agoScore 85

AI agent governancehuman-in-the-loopAWS services

Products Editor selected

百度无人车世界模型实现实时全局推演，一天模拟抵三年

百度无人车团队推出基于世界模型的实时推演系统，能够模拟全局交通场景，包括具有独立行为的行人。该系统一天的计算量相当于传统方法三年的模拟量，大幅提升自动驾驶测试效率。该技术可能加速无人车在复杂环境中的验证与部署。

Why it matters: 该报道展示了百度在自动驾驶世界模型上的具体进展，具有技术突破和行业应用价值，对关注自动驾驶和AI模拟的读者有参考意义。

qbitai23h agoScore 82

自动驾驶世界模型AI模拟

Products

Daybreak Red now available in AWS GovCloud (US-East) via Amazon Bedrock

OpenAI's Daybreak Red, which includes GPT-5.6 Cyber, is now available to approved customers in AWS GovCloud (US-East) through Amazon Bedrock. The offering is restricted to authorized cybersecurity work. This expands access to advanced AI capabilities in a regulated government cloud environment.

AIHUB1d agoScore 78

AI in governmentcybersecuritycloud deployment

Products Editor selected

openJiuwen Agent-Memory: First Audio-Visual Dual-Stream Memory for Agents

openJiuwen introduces Agent-Memory, a system that gives AI agents a dual-stream memory for both audio and visual content, enabling them to understand long videos and recall key events accurately. The system reportedly boosts long-video memory precision and cuts question-answering latency by up to 97%. This marks a significant step in agent memory capabilities for multimodal understanding.

Why it matters: Provides a concrete, recent AI development in agent memory for long-video understanding, with measurable performance gains.

qbitai1d agoScore 82

AI agentsmultimodal learningvideo understanding

Products Editor selected

OpenAI Introduces Images 2.6 with Faster Generation and GIF Creation

OpenAI has released Images 2.6, an upgraded image generation model now available across ChatGPT, Codex, and the API. The update includes faster generation, GIF creation, more precise editing, and new API controls. These features aim to enhance creative workflows and developer integration.

Why it matters: This is a concrete product update from a major AI provider, offering new capabilities that are directly useful to users and developers.

OpenAI News1d agoScore 85

OpenAIimage generationAI tools

Products

datasette-agent 0.5a0 released with new panel feature

Datasette Agent 0.5a0 introduces a panel that lets the agent chat sit alongside Datasette pages like tables or queries. Combined with datasette-apps, users can prompt a database-backed app into existence and modify it via the chat. The release is available for testing by signing in with GitHub at agent.datasette.io.

Simon Willison1d agoScore 75

DatasetteAI agentdatabase-backed apps

Products Editor selected

Anthropic is working on a system to prove AI models' identity

Anthropic is developing FlashProve, a system designed to verify that an AI response genuinely originates from the claimed model. This would allow anyone to check the authenticity of AI outputs, addressing concerns about impersonation and trust. The system is still in development, with details on implementation and release not yet disclosed.

Why it matters: This is a concrete, recent development from a major AI lab addressing a critical trust issue in AI deployment, offering clear reader value for those concerned about AI authenticity.

techcrunch1d agoScore 85

AI authenticitymodel verificationAnthropic

Products Editor selected

Anthropic Releases FlashProve: Cryptographic Proofs That a Reply Came From Committed Weights

Anthropic's Frontier Red Team released FlashProve, an open-source zkSNARK system for provable inference. It proves that a model's reply came from committed weights without revealing them, achieving ~110 tokens/second on a 400B MoE model using one H100. Proofs are 18.78 MB and verify on a CPU in 17 seconds, though proving costs 70-95x the GPU time of standard inference. The system is formalized in Lean and targets use cases like verifying model provenance and detecting tampering.

Why it matters: This is a concrete, recent AI development with significant performance improvements over prior zkML systems)Skip, offering practical value for model verification and security.

MarkTechPost1d agoScore 85

zkMLAI securitymodel provenance