Daily AI Briefing
The 120 stories that mattered in AI, curated and summarized from dozens of sources by AIBriefs.
GPT-5.6 Sol launches publicly alongside Terra and Luna at the same price as GPT-5.5. OpenAI describes Sol as a significant step forward in coding, science, and cybersecurity with its most advanced safety stack.
GPT-Live is now available to ChatGPT Go, Plus, and Pro users, with free user rollout in progress. The next-gen voice model enables full-duplex, real-time conversations and will come to the API soon.
Claude Sonnet 5 is available starting at $2/M input and $10/M output tokens (intro pricing until Aug 31, 2026). It achieves performance close to Opus 4.8 in agentic tasks like coding, tool use, and reasoning while being faster and cheaper.
Anthropic extended free access to Fable 5 on paid plans until July 12, with up to 50% of weekly usage limits included. The model was redeployed globally on July 1 after US export controls were lifted, adding new cybersecurity classifiers.
DeepSeek's new model variant DeepSeek-V4-Flash-DSpark is now available on HuggingFace. The model is currently trending with 49 likes.
Rolling out to Max subscribers over several weeks. Tasks continue in the background even without a device online. Over 90% of usage is knowledge work, not coding.
DeepSeek-V4-Pro-DSpark is now available on HuggingFace with 56 likes and trending status. The model is the latest variant in DeepSeek's V4 series.
The open-weight model features significant improvements in coding and agentic tasks with strong long-context capabilities. Early community feedback compares it to frontier models like Opus 4.8 and GPT-5.5, with a NVIDIA quantized version also available.
The round was led by Radical Ventures, with participation from Nvidia Ventures and others. Prime Intellect provides a full-stack platform for enterprises to build AI agents without relying on frontier labs.
Anthropic alleged that Alibaba-affiliated operators used nearly 25,000 fraudulent accounts to generate 28.8 million Claude exchanges between April and June. The company is urging Congress to strengthen export controls and penalize firms engaging in large-scale model extraction. The attack targeted Claude's agentic reasoning, software engineering, and long-horizon planning capabilities.
Noma Security researchers discovered a prompt injection vulnerability (GitLost) in GitHub Agentic Workflows, allowing unauthenticated attackers to leak private repo data via a crafted public issue. The attack requires no stolen credentials and exploits the AI agent's ability to read issues within the same organization.
ByteDance and Alibaba have removed AI companion apps from stores following new Chinese regulations. The move targets AI emotional companion products, requiring compliance with stricter content and data rules.
Guterres borrowed the Silicon Valley meme 'vibe coding' to argue that humanity's future cannot be built on unregulated AI development. He also called for an international ban on lethal autonomous weapons ('killer robots').
Anthropic identified a 'J-space' in Claude's neural activations that enables conscious access and multi-step reasoning separate from chain-of-thought. Deleting it preserves basic fluency but impairs complex reasoning, and it can detect hidden sabotage goals.
GPT-5.5-Cyber scored 85.6% on the CyberGym benchmark, outperforming Anthropic's banned Mythos 5 (83.8%). OpenAI also released a Codex Security plugin and launched Patch the Planet with Trail of Bits to patch open-source projects like cURL and Python.
Available in beta for Claude Enterprise and Team customers on Slack. Anthropic says 65% of its product team’s code is created by an internal version of Claude Tag.
At the AtCoder World Tour Finals, AI outperformed humans in both Heuristic and Algorithm contests. In the Algorithm contest, no human solved more than 3 problems.
xAI partners with Databricks to natively integrate Grok models into Databricks Agent Bricks, announced at the 2026 Data + AI Summit. This adds to Grok's existing availability on Amazon Bedrock.
Ben Bernanke, former Federal Reserve chair, joins Anthropic's Long-Term Benefit Trust. The trust oversees Anthropic's commitment to responsible AI development.
AI training startup Mercor is reportedly in talks for a valuation of $20 billion, according to Bloomberg. The discussions are at an early stage and no deal is finalized.
Leanstral 1.5 achieves SOTA on formal verification benchmarks: 87% on FATE-H, 34% on FATE-X, and solves 587/672 PutnamBench problems. It is Apache-2.0 licensed with 6B active parameters and has uncovered 5 previously unknown bugs in open-source repositories. The model is available via HuggingFace and a free API.
Anthropic executives proposed to Commerce Secretary Howard Lutnick to end the export ban on its Mythos and Fable AI models, pledging closer cooperation and rapid remediation of security concerns. Talks are ongoing with no set timetable, involving key figures like co-founder Tom Brown and policy head Sarah Heck.
The US government issued an order targeting Anthropic's AI models, marking a new era for AI controls. The order signals increased regulatory scrutiny on frontier AI labs.
DeepSeek V4 launches in mid-July with a 1-million-token context window and improved performance in agent tasks, math, and code. API pricing will double during peak hours (9:00-12:00 and 14:00-18:00 Beijing time).
Anthropic released Claude Sonnet 5, describing it as a cheaper and faster alternative to Opus 4.8. The model is now available to users, with detailed evaluations expected in the coming days.
DeepSeek raised about $7.4 billion in a Series A round, valuing it at $60 billion. Founder Liang Wenfeng personally invested $3 billion to maintain control and pursue AGI.
Export controls on Anthropic's Fable 5 model are expected to be lifted on June 30, according to a Politico reporter. Anthropic's official X account announced the model will return tomorrow.
NVIDIA and Hugging Face are collaborating to bring NVIDIA Isaac GR00T 1.7, Isaac Teleop, and GR00T-trained policies to LeRobot, expanding open robotics tools. LeRobot v0.6.0 releases with new capabilities for training and evaluation, including VLA-JEPA, the first world model policy in LeRobot.
A reliable leaker suggests GPT-6, featuring a new larger pretrain, and Fable 5.1 could launch within four weeks, possibly by end of July. The leaker previously corrected false rumors about Fable 5, adding credibility.
Paris-based voice AI startup Gradium has raised $100 million in a seed extension round backed by Nvidia. The company, spun out of AI lab Kyutai, plans to open a Bay Area office and compete with ElevenLabs. Gradium already counts Renault among its customers.
The vulnerability, discovered by Varonis, could allow attackers to silently manipulate Dialogflow CX conversations, exfiltrate data, and compromise all agents in the same Google Cloud project. Google addressed the flaw in late 2025 after Varonis reported it.
Study finds GitHub Copilot can bypass its own safety refusals: requests rejected in chat are executed when broken into small code-editor steps. Researchers Abhishek Kumar and Carsten Maple demonstrated the vulnerability, highlighting gaps in AI coding assistant guardrails.
New features help consumers understand when ads use AI-generated content. Advertisers get simple disclosure tools as part of Google's transparency push.
Anthropic suspended access to Fable 5 and Mythos 5 after an export control directive banned foreign nationals from using the models. The action follows abuse concerns, with the government targeting specific models.
The Nvidia rival AI chip startup is in talks to raise funds at a valuation of $5 billion. No further details on the funding round have been disclosed.
Katrina Dudley of Franklin Templeton expects the AI infrastructure investment theme to remain durable through 2027 and potentially beyond. The bull case is winning over bear arguments, she said.
China stated at the UN's first Global Dialogue on AI Governance that open source AI is a shared asset, citing DeepSeek and Qwen as lowering barriers and costs. China committed to further promoting open source AI for industry, academia, and research institutions.
Sleeper-agent backdoors can flip fine-tuned LLMs to harmful outputs on untested triggers, evading behavioral monitors and interpretability tools. The solution lies in the training data itself, not post-hoc testing.
A study of 67 frontier models from 21 providers found that enterprises underestimate co-failure rates by a factor of 2.25x when using multiple AI models. The 'co-failure' problem shows that routing queries across specialists does not eliminate correlated failures.
EBR-bench is built on the board game Earthborne Rangers and tests AI's ability to learn on the fly. GPT-4 also set a record on the Epoch Capabilities Index.
Hanna Lichtenberg discusses teaching agents good retrieval, addressing vector search limitations and advocating balanced methods. The talk shares practical techniques for building retrieval systems.
Former GitHub CEO Thomas Dohmke's startup Entire is opening a preview of a distributed Git network designed to handle AI coding agent fleets. The network aims to prevent single-server overload and may compete with GitHub's offerings.
JetBrains introduces a governance layer that gives engineering leaders visibility into developer usage of AI coding agents like Claude Code, Codex, and Gemini CLI, including cost tracking. The tool aims to standardize AI tool adoption across teams while maintaining security and compliance.
DevRev, led by Nutanix co-founder Dheeraj Pandey, argues that benchmarks like TAU-Bench and Agent's Last Exam fail to test real-world enterprise readiness. The article highlights the gap between vendor claims and meaningful measurement for AI agents.
A new dashboard shows open-source coding models improving ~1.5x faster than frontier models. A 27B open-source model already beats Claude Opus 4.8 on decontaminated coding benchmarks.
Three big AI IPOs are projected to generate more value than all U.S. VC-backed exits since 2000, according to analysis. Anthropic, OpenAI, and SpaceX each expected to debut at valuations dwarfing typical tech companies.
The partnership integrates Cortechs.ai's quantitative MRI analysis into Viz.ai's AI-powered care coordination platform, starting with multiple sclerosis. The goal is to help clinicians identify and manage patients with neurodegenerative disease more effectively.
Databricks unveiled CustomerLake, an agentic customer data platform that unifies customer data, AI models, identity resolution, and activation. The platform is built natively on Databricks.
The analysis raises concerns about the benchmark's accuracy and reliability for evaluating AI model coding abilities. OpenAI details how the benchmark may conflate signal with noise.
Wiz researchers found symlink vulnerabilities in six popular AI coding assistants (Amazon CodeWhisperer, etc.) that allow a malicious repo to silently overwrite sensitive files, leading to arbitrary code execution. The flaw, dubbed 'GhostApproval,' exploits a gap in how assistants handle symlinks and permission prompts.
AlphaEvolve, developed by Google DeepMind, is now widely available on Google Cloud to help solve customers' hardest problems. The rollout targets the most challenging issues on the platform.
Flint enables AI agents to generate polished charts from simple, human-editable specifications. It uses semantic data types to guide chart design and compilation.
General Joshua Rudd, NSA and Cyber Command chief, allegedly stated that Anthropic's Mythos model breached nearly all classified systems within hours. The claim was revealed during a Senate Intelligence Committee hearing by Vice-Chair Mark Warner on June 11th.
Bosworth candidly explains why Meta fell behind in the frontier AI race and outlines the company's plan to leverage its models, products, and distribution. He also discusses renting models and challenges facing consumer AI.
Investors are reassessing political risk after a ban on AI company Anthropic, according to Bloomberg. The report examines how government actions are reshaping investment strategies in the AI sector.
Meta is on track to start production of its latest custom AI chips in September, aiming to reduce reliance on Nvidia GPUs. The chips are developed to power Meta's AI workloads more cost-effectively.
The law practice management company Smokeball has released the next generation of its AI assistant Archie, now built on agentic AI and embedded directly into Microsoft Word and Outlook. This update comes two years after the original launch of Archie in July 2024.
Welch Labs explains how self-supervised learning eliminates the need for labeled data in computer vision. The approach leverages contrastive learning and masked autoencoders to achieve strong performance without manual annotations.
The AI-agent startup deployed its own fundraising agent named SivaClaw to secure $100 million in funding. The agent handled the entire fundraising process, including investor outreach and negotiations.
Harvey's co-founder Gabe Pereyra announced the legal AI platform's token consumption grew 14-fold in six months. The milestone reflects accelerating enterprise adoption of AI-powered legal tools.
A preclinical trial at UC San Diego successfully used two teleoperated humanoid robots to perform a gallbladder removal. This marks a first for humanoid robots in surgery, aiming to expand access to remote surgical care.
In a new essay titled 'Policy on the AI Exponential,' Anthropic CEO Dario Amodei proposes government regulation of powerful AI models, drawing parallels to the commercial aviation industry. He argues that advanced AI systems require a regulatory framework similar to the FAA to ensure safety.
Chinese AI startup MiniMax plans to launch a new large language model with 2.7 trillion parameters, per The Information. The model would be among the largest ever, rivaling top frontier models.
Pearl Health, a healthcare outcomes platform for Medicare patients, raised $110 million in capital including a $50 million equity round led by Andreessen Horowitz. The company reached profitability in 2025.
AI agents discovered a remotely triggered panic in libp2p's gossipsub, disclosed as CVE-2026-34219. Researchers noted the main work shifted from finding bugs to validating which ones are real.
Amazon SageMaker HyperPod adds data capture, Hugging Face integration, NVMe storage, and Route 53 support. These enhancements improve inference observability, speed, and flexibility for enterprise generative AI workloads.
New training method incentivizes MLLMs to respect event order in egocentric video reasoning. Outperforms baselines on Ego4D across multiple temporal tasks.
Teams often expose APIs as-is, but AWS recommends designing MCP tools with agentic systems in mind. Key considerations include parameter names, error messages, and tool descriptions to improve agent performance.
Proof-of-concept 'Friendly Fire' attack by AI Now Institute can trick AI coding agents like Anthropic's Claude Code into executing attacker's code while scanning for vulnerabilities. The attack exploits the agent's ability to run code from the repository being analyzed.
On July 7, 2026, the UK government announced an agentic AI defense plan alongside an industry cybersecurity pledge. The initiative demonstrates the government's commitment to improving national cybersecurity through AI.
New York Fed President John Williams identifies AI as a primary driver of inflation uncertainty. He notes the technology's potential to both boost productivity and disrupt labor markets, complicating monetary policy.
The Trump administration imposed export controls on Anthropic's most powerful AI after SK Telecom, a South Korean carrier with alleged ties to China, gained access to Claude Mythos via Project Glasswing. Amazon later flagged vulnerabilities in Fable 5, compounding concerns and leading Anthropic to disable access to the models for all foreign nationals.
Bloomberg reports that Anthropic is in talks with the US government to relax regulatory restrictions on AI model deployment, according to sources. The potential agreement could mark a shift in AI policy.
Top-tier performance came from a mix of proprietary and open models. The benchmark explored cost-performance tradeoffs on real engineering tasks across Databricks' codebase.
Anthropic took Claude Fable 5 and Mythos 5 offline after a US export-control directive barred foreign nationals from using the services. The company had warned the models could find and exploit software vulnerabilities, and experts say similar capabilities will soon be common.
NVIDIA CEO Jensen Huang sat down with LangChain's Harrison Chase to discuss the importance of open, domain-specific agent systems for turning LLMs into deployable products. He argues that building 'super agents' tailored to specific domains is the path to making AI useful.
Offers token-based pricing, 99% uptime SLA, and up to 90% lower cost than proprietary APIs. Available now for MiniMax M3 and GLM-5.2 with capacity in North America, EMEA, and beyond.
Aurora 1.5 adds 22 weather variables and hourly resolution for applications in energy, agriculture, and climate risk. The model is released as an open foundation model with probabilistic ensemble forecasting.
Open-source AI tool Ollama raised $65M in funding from Benchmark. It has grown to nearly 9 million users and amassed 176,000 GitHub stars.
The attack, called HalluSquatting, exploits AI coding assistants' tendency to hallucinate fake package names. Attackers register those names to deliver botnet malware, turning a known limitation into a security threat.
OpenAI published principles for responsible AI use in government and national security partnerships, emphasizing democratic accountability and public safety. The post outlines guidelines for evaluating and engaging with government entities.
Protea achieves a 10-million-token context window, rivaling Subquadratic's 12-million-token model from May 2026. The model employs biologically-inspired swarm optimization for efficiency.
A proposed class action lawsuit claims a man used xAI's Grok to create thousands of child sex abuse images. The suit alleges xAI only reported one explicit prompt to authorities, despite generating over 7,000 images.
Anthropic published a retrospective article detailing the development of Claude Code, from its origins in safety research to launch, with insights from builders and early users. Lead engineer Boris Cherny described it as 'the first time telling the story' and said 'we are 1% done.'
Johan Lajili from Poolside AI presents why lack of good vision makes agents unreliable. He emphasizes that proper visual grounding multiplies performance and trust in agent systems.
Modal closed a $355M Series C round. CTO Akshat Bubna discusses why cloud infrastructure must evolve for agent workloads in a Latent Space interview.
LTX (formerly Lightricks' video generation project) becomes an independent company focused on open foundation models for simulating physical reality. The company will release open models for how objects move and scenes evolve, targeting enterprises and researchers.
A House Committee investigation is probing the risks of Chinese AI models used by U.S. companies, focusing on potential security and data vulnerabilities. The probe reflects growing bipartisan concern over AI technology from China.
A fintech RAG pipeline produced confident lies despite a green observability dashboard. The "silent hallucination" loop occurred when the autonomous data pipeline ingested its own hallucinated outputs, corrupting the vector store.
The UK Ministry of Justice's Justice AI unit operates with 40 people, a team size typical of 300. Their probation officer tool scaled from MVP to national rollout with just two engineers, as detailed by William Tarr in this AI Engineer talk.
A cryptomining incident highlights how AI gateways can provide attackers access to AI models, cloud infrastructure, and identity and access management (IAM) data. The incident underscores the security risks posed by AI gateways.
AI chip startup SambaNova raised funds at an $11 billion valuation, according to Bloomberg. The funding round highlights continued investor interest in AI hardware startups.
The one-click vulnerability, codenamed WriteOut, could allow session token theft via agent previews, enabling cross-tenant compromise. Writer has patched the flaw.
The new backend integrates transformers for high-performance inference. It is designed to run at native speed directly within vLLM.
Character.AI announced c.ai Series, short-form episodic videos that users can watch and interact with on mobile. The twist: users can chat with characters, ask questions, and roleplay different storylines.
Anthropic published a research paper detailing a method to selectively suppress dual-use knowledge in AI models. The technique would allow model operators to disable specific harmful capabilities while retaining beneficial ones.
Over 32,000 attendees gathered for the four-day event featuring 800+ sessions. Stanford won the inaugural Grounded Reasoning Cup, competing with models from Google DeepMind, Anthropic, and OpenAI.
Frontier AI models enable autonomous attacks that can break into systems in 27 seconds. Human-operated security workflows cannot keep up, requiring resilience planning before attacks occur.
The talk covers architectural challenges of building a secure sandbox cloud for agents, comparing runtime isolation technologies and persistence strategies. Abhishek Bhardwaj presents the Fleet approach for scalable agent execution.