Latest AI News

SpaceX completed its $60 billion acquisition of coding startup Cursor on Friday in an all-stock transaction. The deal was announced as an option in April as part of a partnership agreement that allowed Cursor to use SpaceX resources. The acquisition marks a significant move by SpaceX into the AI-powered developer tools space.

French startup Kog challenges the assumption that GPUs are poorly suited for agentic AI workflows. The company is developing deeper optimization techniques to extract more inference performance from GPU hardware. This work suggests that current GPU utilization for agent-based AI tasks may be suboptimal rather than fundamentally limited by hardware design.

Tarun Chhabra, Anthropic's head of national security policy and a former Biden administration official, is transitioning out of his current role. He will move into a new advisory position at the AI company focused on geopolitics and national security policy. The move reflects shifts in how Anthropic is structuring its policy and government relations work.

IBM has partnered with OpenAI to expand enterprise AI adoption, with plans to train and certify tens of thousands of consultants on OpenAI's technologies. The deal positions IBM to deepen its consulting services around generative AI for corporate clients. This move reflects intensifying competition among enterprise software vendors to build AI-native service capabilities.

Apple has developed a custom large language model for the Chinese market in partnership with Alibaba, marking a shift from its previous reliance on domestic Chinese AI solutions. The collaboration represents a rare cross-border tech partnership amid escalating US-China tensions. The move gives Apple greater control over its AI products in China's competitive smartphone market.

OpenAI has appointed Dali Rajic, president and chief operating officer of Alphabet-owned cybersecurity firm Wiz, as its new chief revenue officer. Rajic replaces Denise Dresser, who is departing to pursue other opportunities after a transition period. The move signals OpenAI's focus on scaling revenue operations as the company continues to expand its commercial offerings.

DeepSeek launched DeepSeek-V4-Pro, an updated flagship model for agentic workloads, alongside DeepSeek Harness v0.1, an open-source agent framework available under MIT license. The releases position DeepSeek as a competitor to Anthropic's Claude Code and OpenAI's Codex by offering developers an alternative agent infrastructure layer. Simultaneously, DeepSeek is shifting from flat API pricing to peak and off-peak rates starting August 16, with substantially higher prices across the board.

OpenAI has launched Ultrafast, a new API service tier that runs GPT-5.6 Sol at speeds up to 14 times faster than standard offerings, powered by Cerebras infrastructure. The service delivers up to 750 output tokens per second. This represents a significant acceleration in inference speed for enterprise and developer users requiring real-time or near-real-time AI responses.

Chinese AI developer Z.ai released GLM-5.3, an open-source model it claims matches Anthropic's Mythos 5 in cybersecurity capabilities. The Beijing-based company, also known as Zhipu, positioned the model as a significant improvement over its predecessor GLM-5.2. The release marks another step in China's competitive push in generative AI development.

Google DeepMind released Gemini 3.7 Flash on August 13, 2026, positioning it as an improved workhorse model for coding and agent-based tasks. The model arrives three weeks after Gemini 3.6 Flash and delivers measurable gains in software engineering, web development, and knowledge-intensive workflows at half the per-token cost of its predecessor. The release reflects developer feedback and algorithmic improvements aimed at production-ready code generation and complex document processing.

AI training companies like Mercor are actively acquiring internal communications and code from startups, offering payments up to $300,000 for Slack threads, GitHub records, and meeting transcripts. Warmly's CEO received four such acquisition offers within days of the company's HubSpot acquisition announcement. The practice highlights how internal startup data has become a commodity for AI model training, even as acquirers may not want the same datasets.

Capital One built a multi-agent AI platform centered on customized open-weight models rather than relying on off-the-shelf frontier models. The bank fine-tunes open models with proprietary data and uses a specialized multi-agent orchestration system called MACAW to handle complex workflows like fraud detection and customer service. This approach leverages Capital One's data advantage while enabling extensibility across the enterprise.

AI chip shortages are creating acute pressure on startups building proprietary models, forcing founders to negotiate constantly with cloud providers for GPU access. Evan Morikawa of Generalist, which trains AI models for robotics, recently contacted 17 different providers to secure compute capacity. The scarcity has made GPU pricing and availability a critical business variable for model-training startups with limited funding.

Microsoft is consolidating its consumer and commercial Copilot AI assistants into a single unified app called Microsoft Copilot, eliminating the need for separate applications in the taskbar. Both personal and work accounts will migrate to the new interface, which combines chat, image creation, and Microsoft 365 integration. The move reduces UI clutter while centralizing access to Copilot functionality across consumer and enterprise use cases.

A MIT Technology Review Insights report based on a survey of 300 data and technology executives finds that legacy data systems are a major blocker to AI agent adoption and effectiveness. Organizations with mature data infrastructure, termed 'data leaders,' report significantly higher trust in agent decisions and fewer scaling constraints than 'data laggards.' The research suggests that without modernizing data systems, enterprises will struggle to realize ROI from agentic AI despite widespread adoption plans.

Nvidia-backed cloud compute firms CoreWeave and Nebius are capitalizing on surging demand for AI chip capacity by raising prices sharply. Nebius held its first computing capacity auction in Q2 with Blackwell chip prices 15% above previous highs, and is now selling capacity closer to delivery dates to exploit price volatility. The trend reflects intense competition for limited Nvidia GPU supply among AI companies.

Geoffrey Hinton, Fei-Fei Li, and Andrew Ng debated AI regulation, open source access, and U.S. competitiveness against China at the Ai4 conference. The three pioneers discussed how America can maintain its position as China advances in Asia, with particular focus on whether open source development serves or undermines safety and national interests. Their discussion reflects a core tension in AI policy: balancing innovation speed with safety oversight and geopolitical competition.

Thrive Holdings, an OpenAI-backed enterprise AI company, has raised $2 billion in Series funding at a $12 billion valuation. The round was led by investors including SoftBank, D1 Capital Partners, and Altimeter Capital. The funding reflects growing investor appetite for AI applications targeting business customers rather than consumer markets.

SpaceXAI released Grok 4.6, scoring 61 on Artificial Analysis Intelligence Index and tying OpenAI's GPT-5.6 Sol for third place globally. The model targets long-running agents, coding, and knowledge work with pricing starting at $2 per million input tokens and $6 per million output tokens, less than half the cost of GPT-5.6 Sol standard mode. The release emphasizes improvements in agent behavior and task persistence rather than isolated benchmark gains.

Cisco Systems reported strong fourth quarter revenue growth driven by cloud providers increasing spending on AI networking chips and switches, yet shares fell 5% following the earnings announcement. The decline suggests investor concerns about valuation or forward guidance despite the positive demand signals from major cloud customers. The company's AI networking portfolio is becoming a material revenue driver as cloud infrastructure providers scale AI deployments.

Cognition, an AI coding startup, is reportedly in early-stage talks to raise capital at a $40 billion valuation, just months after closing a $1 billion funding round at a $26 billion valuation. The move signals continued investor appetite for AI-powered developer tools despite recent market volatility. The company's rapid valuation increase reflects the competitive pressure in the AI coding space and investor confidence in its technology.

Lovable, an AI-powered software development startup, confirmed a new $13.3 billion valuation and raised $400 million in fresh funding. The round comes after the company achieved $500 million in annualized run rate revenue in June. The funding reflects investor confidence in AI-assisted coding tools as a category.

Google DeepMind has introduced sign-language-to-text (SL2T), a new AI model that converts sign language into text for Deaf and hard of hearing users. The model powers new sign language features designed to improve accessibility. The announcement marks a significant step in making AI tools more inclusive for sign language users.

Tencent nearly tripled its capital expenditures to 52.8 billion yuan ($7.8 billion) in Q2 2026, driven by investments in computing infrastructure for AI model training and its coding and productivity tools. The spending surge reflects the Chinese tech giant's commitment to scaling AI capabilities amid intensifying competition in the sector. The capex increase signals Tencent's pivot toward infrastructure-heavy AI development.

Blacksmith, an AI code-testing startup, has seen its valuation jump nearly 10x to $550 million in less than a year, driven by tenfold revenue growth. The company operates in the software validation space, where AI-powered testing tools are gaining traction among developers and enterprises. The valuation surge reflects investor appetite for AI-native developer tools amid broader adoption of AI in software development workflows.

Skan AI raised $63 million in Series C funding to build what it calls a 'context graph of work' by observing how employees actually perform jobs across enterprise software. The company is launching two new products, Skan AI Blueprint and Skan AI Agents, alongside its existing intelligence offering. The funding and product announcements address a widespread problem in enterprise AI, where only 8% of companies have AI agents in production and 95% of early implementations require complete redesign.

The U.S. Department of Agriculture has reduced its Salesforce footprint by letting a contract expire and shifting correspondence management to C3 AI's software, deployed through reseller Carahsoft. The move, led by new CIO Sam Berry, replaced a $6 million annual Salesforce contract with a $9 million one-year C3 contract that used AI to clear a backlog of roughly 2,000 unanswered letters in two months. The shift reflects broader cost optimization efforts within the Trump administration's DOGE initiative.

OpenAI has released a dedicated ChatGPT desktop application for Linux, extending its native app availability beyond Windows and macOS. The move addresses a gap in platform coverage for Linux users who previously lacked a first-party desktop client. This represents OpenAI's continued effort to broaden ChatGPT's accessibility across operating systems.

River AI, a startup founded by xAI co-founder Igor Babuschkin, raised $1.1 billion in funding led by General Catalyst just two months after launch. The company is developing personal agents, though specific details about its technology and product roadmap remain limited. The funding round signals investor confidence in the personal AI agent space despite the startup's early stage.

Anthropic's unreleased AI model has made measurable progress on the Riemann hypothesis, one of mathematics' most significant unsolved problems that has resisted solution for over 150 years. The company has not solved the problem, but the model's progress exceeds typical expectations for AI applied to such fundamental mathematical challenges. The development signals growing capability of large language models in tackling complex mathematical reasoning.

OpenAI and AWS have integrated Daybreak cybersecurity capabilities into Amazon Bedrock, making the models available to enterprise customers. Daybreak is positioned to support security workflows through AWS's managed service for foundation models. The partnership expands access to OpenAI's cybersecurity-focused AI tools for organizations already using AWS infrastructure.

Junyang Lin, a former Alibaba researcher who departed in March, has launched Pragmatik Labs, a Shanghai-based startup focused on next-generation AI agents for both digital and physical applications. The announcement confirms earlier reporting and marks Lin's entry into the competitive AI agent space. The startup's focus on agents spanning digital and physical domains positions it within a growing segment of AI development.

SpaceXAI, the rebranded xAI division, is launching Grok Bot in beta, an AI agent designed to operate continuously across enterprise software applications rather than respond to individual prompts. The product costs $120 per seat per month for teams and $200 for individuals, positioning it as a persistent digital coworker that can execute workflows autonomously. The launch arrives as Anthropic, OpenAI, and other competitors expand their own agentic capabilities for enterprise use.

Kalshi, a prediction market platform, has reached an annualized revenue run rate exceeding $4 billion in July, driven by World Cup wagering activity. The company is in advanced talks to raise funding at a $40 billion valuation, nearly double its May valuation. Growth prospects are tempered by heavy marketing spending and potential new tax obligations in major markets.

Brad Lightcap, OpenAI's special projects lead and former COO, is departing after eight years at the company. In an internal memo posted to X, Lightcap said he would be starting something new and believes there are important things the world will need to get right in the next period. He indicated he would remain available for the next few weeks during the transition.

Google's Gemini app has reached 1 billion monthly active users, according to CEO Sundar Pichai, making it the fastest-growing product in the company's history. The milestone signals Google's growing competitive position in the consumer chatbot market, which has been dominated by OpenAI's ChatGPT. The announcement was made Tuesday on X.

OpenAI has solved 10 long-standing mathematics problems, some unsolved for decades, using AI technology that identifies patterns across vast datasets. The breakthrough is prompting leading mathematicians, including Fields Medal winner James Maynard at Oxford, to reassess the future of their discipline as mathematics adapts to AI capabilities. The development signals that generative AI, already transformative in text, images, and scientific research, is now reshaping how mathematical problems are approached and solved.

Joby Aviation is acquiring Dayton, Ohio-based defense contractor Resonant Sciences for $500 million, financed through $450 million in cash and $50 million in equity. The deal, expected to close in early 2027, positions Resonant as Joby's dedicated defense division while the company's commercial air taxi business continues separately. The acquisition reflects Joby's strategic expansion into military markets alongside its consumer aviation ambitions.

Anthropic has committed to embedding machine-readable watermarks in Claude-generated text and images to comply with European AI transparency regulations. The watermarks will be invisible to humans but detectable by people and platforms seeking to identify AI-generated content. The company says the changes are a future commitment rather than an immediate rollout, and will include digitally signed provenance metadata where supported.

Manus, an AI startup acquired by Meta, announced Tuesday it will return to operating as an independent company following a Chinese government order in April to unwind the deal. The reversal marks a significant setback for Meta's AI acquisition strategy and reflects growing regulatory scrutiny of foreign tech acquisitions in China. The company said it is close to finalizing the separation.

LTX released LTX-2.5, an open-weights video generation model that produces 10-second clips in 6.8 seconds on Nvidia GB200 chips, with native multishot support and improved quality. The model is available free for organizations under $10 million ARR on Hugging Face, ComfyUI, and via API. LTX claims 33 million downloads across its model family and reports a 67% win rate in blind quality tests against competing models.

Mistral AI announced a three-part infrastructure expansion targeting 1 gigawatt of European compute capacity by 2030, starting from less than 200 megawatts today. The plan includes regional inference endpoints, a new Priority Tier with uptime guarantees, and a coalition of European enterprises making multi-year compute commitments to fund 200 megawatts by end of 2027. The company is also hosting third-party open models, including GLM-5.2 from Chinese AI lab Z.ai, marking a shift from open-weight model training toward critical infrastructure services.

NVIDIA released Nemotron 3.5 Lightning, a 30-billion-parameter mixture-of-experts model designed for specialized tasks in multi-agent AI systems, alongside NeMo Switchyard, an open source routing library. The model delivers up to 4x faster output speed and 30% faster agentic task completion compared to competitors in its class. Both tools enable enterprises to deploy customized AI across local systems, edge devices, and cloud infrastructure without rewriting applications.

Spotify is implementing 'AI Persona' labels for artist profiles representing AI-generated identities and will exclude their music from editorial, algorithmic, and personalized recommendations by default. The move addresses the growing presence of synthetic artists on the platform while maintaining their availability for direct search and discovery. This represents a platform-level approach to managing AI-generated content visibility in music streaming.

Meta released Muse Glimmer, a 30-billion-parameter open-weight AI model licensed under Apache 2.0, designed to run autonomous agents on consumer hardware like high-end Macs and PCs. The release marks Meta's return to fully open source after shifting to proprietary models in April, and comes with fewer restrictions than Meta's previous Llama family. Meta also announced plans to open-source Muse Spark 1.2, its frontier model powering the recently launched Muse Code terminal agent.

AWS announced at Black Hat USA 2026 that its Continuum vulnerability platform will integrate directly into Anthropic's Claude Code and OpenAI's Codex, embedding AWS security tooling at the point where developers write code regardless of which AI model they use. The move positions AWS as a security control plane for enterprise software development and reflects an urgent industry response to frontier AI models like Claude Mythos Preview, which identified thousands of previously unknown zero-day vulnerabilities during testing. AWS also expanded its Security Hub Extended marketplace with a 10th category focused on supply chain protection, adding Chainguard and Socket as partners.

Mark Zuckerberg republished a 6,500-word essay on Meta.com that largely duplicated a 1,000-word Wall Street Journal opinion piece published two weeks earlier, both arguing that AI's future should be accessible to everyone. The piece is part of a broader trend of lengthy essays from AI leaders including Dario Amodei at Anthropic, Sam Altman at OpenAI, and Satya Nadella at Microsoft, who use their platforms to discuss AI and society. The author notes these essays are often long and dry, raising questions about whether they are designed to demonstrate AI's summarization capabilities to consumers.

Anthropic has partnered with Macquarie Asset Management and Singapore's GIC sovereign wealth fund to form a new entity that will develop, operate, and lease AI data centers. The partnership aims to support demand for Anthropic's Claude models, with initial focus on U.S. data centers. This move signals Anthropic's commitment to securing dedicated infrastructure as AI model demand grows.

Caitlin Kalinowski, former head of robotics at OpenAI, has joined Anthropic as a member of technical staff focused on research. The hire signals Anthropic's continued investment in robotics capabilities, following the company's release of robotics research last month. Kalinowski's move represents a notable talent shift between two of the leading AI research organizations.

OpenAI has released GPT-5.6-Cyber, a cybersecurity-focused model available through Daybreak Red for authorized vulnerability research, exploit validation, and security testing. The model is designed to support authorized security professionals in identifying and validating vulnerabilities. The release reflects growing demand for AI tools tailored to defensive security work.