AI News Archive: August 13, 2026 — Part 1
Sourced from 500+ daily AI sources, scored by relevance.
- Anthropic reportedly plans a $2 trillion IPO in October—the largest ever—that will eclipse SpaceX
Anthropic reportedly plans a $2 trillion IPO in October—the largest ever—that will eclipse SpaceX Fortune
Score: 90💰 MoneyAug 13, 2026https://fortune.com/2026/08/13/anthropic-ipo-2-trillion-october-largest-ever-spacex/ - OpenAI’s Revenue Run Rate Tops $40 Billion Ahead of IPO
OpenAI is on track to generate annualized revenue of more than $40 billion based on its current performance, according to people familiar with the matter, roughly doubling its run rate from the end of 2025 and bolstering the company’s plans for a Wall Street debut.
Score: 89💰 MoneyAug 13, 2026https://www.bloomberg.com/news/articles/2026-08-13/openai-s-revenue-run-rate-tops-40-billion-ahead-of-ipo - Elon Musk says xAI will increase data center capacity 7x by 2027 — targeting 10 gigawatts of compute, up to $500 billion in revenue by the end of next year
Elon Musk expects xAI to increase its nameplate power draw to 10 GW by late 2027, which will increase its performance by orders of magnitude what is available to AI today.
- Novel AI model accurately detects key gene mutations and predicts biomarkers across 32 cancer types
Novel AI model accurately detects key gene mutations and predicts biomarkers across 32 cancer types EurekAlert!
- China’s Capital Flight Is Fueling Global AI. Beijing Cracks Down.
China’s Capital Flight Is Fueling Global AI. Beijing Cracks Down. Barron's
- Databricks wanted to raise $1B, investors wanted $15B. It settled on $5B at a $190B valuation.
AI is expensive, Ali Ghodsi tells TechCrunch. With so many investors wanting into his latest round, he said yes to more than planned.
- Pixel 11 AI Features: Google’s Biggest Gemini Push Yet
Google’s Pixel 11 adds proactive Gemini tools, live translation and AI camera features. Here’s what buyers should know about limits and pricing. The post Pixel 11 AI Features: Google’s Biggest Gemini Push Yet appeared first on TechRepublic .
Score: 84🌐 MovesAug 13, 2026https://www.techrepublic.com/article/news-google-pixel-11-ai-features-gemini/ - OpenAI introduces ‘Ultrafast,’ a new mode that makes GPT-5.6 Sol work at 14x the speed
OpenAI is launching a preview of a sped up version of its latest, most powerful model, in an effort to court enterprise users.
- Teaching AI the biology of antibodies speeds drug discovery
Teaching AI the biology of antibodies speeds drug discovery EurekAlert!
- There’s a Fatty Liver Epidemic. AI Could Help Get Ahead of It
Over a billion people worldwide have livers with excess fat, which can lead to a host of medical problems. Researchers think AI tools can spot the condition—and help stop it—early enough to save lives.
- Lovable's $13.3 billion valuation puts AI coding's SaaS challenge in focus
Since launching in November 2024, users have created more than 60 million projects on the platform, which now sees more than 900 million monthly visits to apps built on it
- Thrive Holdings, A.I.-Focused Buyer of Service Firms, Raises $2 Billion
The company, Thrive Holdings, acquires businesses to infuse them with artificial intelligence and drew interest from backers like SoftBank.
Score: 80💰 MoneyAug 13, 2026https://www.nytimes.com/2026/08/12/business/dealbook/thrive-holdings-ai-funding.html - OpenAI Foundation gives $100 million fund state AI implementation for public health
The organization announced the launch of the Breakthroughs to Follow-Through Initiative, which will implement artificial intelligence tools in public health research work.
- Inflation Softens, Nvidia's $500B Plan | Real Yield 8/13/2026
"Bloomberg Real Yield" highlights the market-moving news you need to know. Today's guests: Vanguard Head of Fixed Income Client Portfolio MGMT Matt Wrzesniewsky, BlackRock Chief Investment & Portfolio Strategist, Americas Gargi Chauduri, Nuveen Head of Fixed Income Strategy Tony Rodriguez, and TCW Generalist Portfolio Manager Jerry Cudzil. (Source: Bloomberg)
Score: 80💰 MoneyAug 13, 2026https://www.bloomberg.com/news/videos/2026-08-13/real-yield-8-13-2026-video - Departing Google chief scientist Jeff Dean has been in talks for a $10 billion valuation for his new AI startup
Departing Google chief scientist Jeff Dean has been in talks for a $10 billion valuation for his new AI startup Business Insider
Score: 80💰 MoneyAug 13, 2026https://www.businessinsider.com/former-google-exec-jeff-dean-valuation-for-new-ai-startup-2026-8 - Anthropic could be worth $2 trillion when it goes public
Rapid revenue growth fuels hope Claude maker's IPO is the biggest listing in history
Score: 80🌐 MovesAug 13, 2026https://arstechnica.com/ai/2026/08/anthropic-could-be-worth-2-trillion-when-it-goes-public/ - Anthropic says its own AI models breached three companies during security tests
Anthropic said Thursday that an internal investigation uncovered three incidents in which its AI model Claude breached the systems of three organizations while conducting cybersecurity tests. The investigation, and disclosure, comes more th ... (https://incidentdatabase.ai/cite/1627#7688)
- Google announces Gemini 3.7 Flash just three weeks after previous release
Gemini 3.6 Flash debuted just 3 weeks ago, but Google says 3.7 has "substantial improvements."
Score: 80🤖 ModelsAug 13, 2026https://arstechnica.com/ai/2026/08/google-announces-gemini-3-7-flash-just-three-weeks-after-previous-release/ - UAE's MGX co-leads Databricks strategic funding round
UAE's MGX co-leads Databricks strategic funding round thenationalnews.com
Score: 80💰 MoneyAug 13, 2026https://www.thenationalnews.com/future/technology/2026/08/13/mgx-uae-databricks-ai/ - Nvidia Credit Risk Eases, Still Elevated After $500B Plan
Bloomberg's James Crombie joins Emily Graffeo on "Real Yield." Bond traders dialed back measures of credit risk associated with Nvidia after the company said it would limit its exposure in a $500 billion plan to finance the type of AI investments that are driving demand for its computer chips. (Source: Bloomberg)
Score: 80💰 MoneyAug 13, 2026https://www.bloomberg.com/news/videos/2026-08-13/nvidia-credit-risk-eases-still-elevated-after-plan-video - Lovable bolsters its AI software creation capacity, touts $400M funding round
Lovable, the Swedish-based AI software creation platform company, today announced an acceleration of its product, infrastructure, and team development efforts — and a noteworthy round of Series C funding totaling $400 million . The company said in its statement that it is looking to augment its platform, which it boasts is already used by almost two-thirds of Fortune 500 companies. Headquartered in Stockholm, Lovable plans to grow its team to 450 people this year, hiring most heavily in machine learning, product, infrastructure, and security roles, and is looking to expand operations from its home base to include locations in London and three US cities: Boston, San Francisco, and New York City. The planned growth is made possible by the latest infusion of venture capital, which follows a $330 million Series B funding round last December. Lovable now has a $13.3 billion valuation. The Series C funding was led by Menlo Ventures and the Scaleup Europe Fund and includes US-based Regent. (Regent is the parent company of Foundry.) Earlier this month, Lovable announced a partnership with Cerebras Systems , the chipmaker that builds processors the size of dinner plates (the Wafer-Scale Engine) and supercomputing systems built for AI inference and training. Cerebras systems are designed to keep an entire AI model’s weights on a single, super-sized WSE chip, rather than split across many GPUs, to avoid GPU memory bottlenecks. That partnership calls for Lovable to run some of its latency-sensitive workloads on dedicated Cerebras capacity. “Fast AI is more valuable than slow AI,” said Cerebras CEO and Cofounder Andrew Feldman said in a statement when the partnership was unveiled. “When AI responds in real-time, users do more with it, stay longer, and run higher value workloads. Software creation is one of the clearest examples of the importance of speed. Creators don’t want to wait.” In its statement today, Lovable said that since its launch in November 2024, people have created more than 60 million projects with its tools, with Lovable-built apps seeing more than 900 million visits a month. Computerworld is part of Foundry, which is owned by Regent.
- AI startup Cognition in new funding talks at $40 billion value
Cognition would be expected to raise over $1 billion in the round, Bloomberg reported. The company may still decide not to raise capital, or seek financing with different terms. Founded in 2023, Cognition is backed by top venture firms including General Catalyst and Peter Thiel’s Founders Fund.
- CHAI forms Health AI Cybersecurity Work Group
CHAI forms Health AI Cybersecurity Work Group Healthcare IT News
Score: 80🌐 MovesAug 13, 2026https://www.healthcareitnews.com/news/chai-forms-health-ai-cybersecurity-work-group - The builder’s guide to GPT‑5.6
Learn how startups use GPT-5.6 to build faster, more cost-efficient AI agents with smarter model selection and new Responses API capabilities.
- Gemini 3.7 Flash now available on AI Gateway for 50% off
Gemini 3.7 Flash from Google is now available on AI Gateway for 50% off till December 31st, 2026. Gemini 3.7 Flash improves on prior Flash models at software engineering and agentic work. It resolves issues more reliably and spends less time stuck in failed agent loops, which matters on long tool-calling sequences where one derailment costs the rest of the run. It also generates desktop and web application code directly from design mocks, with closer adherence to the source design. To use Gemini 3.7 Flash, set model to google/gemini-3.7-flash in the AI SDK : To use it in a coding agent , run vercel ai-gateway coding-agents setup to connect Claude Code, Codex, OpenCode, or Pi, then select google/gemini-3.7-flash inside the agent. To try Gemini 3.7 Flash with no code, try the model in the model playground . AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting , Zero Data Retention support , budgets for API keys , routing rules , and more. AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests. Read more
Score: 80🤖 ModelsAug 13, 2026https://vercel.com/changelog/gemini-3-7-flash-now-available-on-ai-gateway-for-50-off - OpenAI previews ‘Ultrafast’ GPT-5.6 Sol running up to 14 times faster
OpenAI is previewing a new way to run its most capable GPT-5.6 model at dramatically higher speeds. The company says its new Ultrafast service tier can run GPT-5.6 Sol up to 14 times faster than standard processing.
Score: 80🤖 ModelsAug 13, 2026https://9to5mac.com/2026/08/13/openai-previews-ultrafast-gpt-5-6-sol-running-up-to-14-times-faster/ - Waymo Imports 3,200 Zeekr Robotaxis Despite Super High Tariffs
Chinese electric vehicles are basically all but banned in the United States. Nonetheless, Waymo has brought in a few thousand Zeekr-based Ojai electric robotaxis. There are 102.5% tariffs on Chinese produced EVs, including these Ojai robotaxis. However, these were custom designed for Waymo’s robotaxi service over several years. So, tariffs ... [continued] The post Waymo Imports 3,200 Zeekr Robotaxis Despite Super High Tariffs appeared first on CleanTechnica .
Score: 80🌐 MovesAug 13, 2026https://cleantechnica.com/2026/08/12/waymo-imports-3200-zeekr-robotaxis-despite-super-high-tariffs/ - Meet Microsoft’s MAI-Thinking-1: What It Is and How It Works
Today Microsoft is introducing MAI-Thinking-1, Microsoft AI’s reasoning model. It is a medium-sized model that stands among the strongest models in its weight class. It matches leading models on key software engineering benchmarks, demonstrates advanced mathematical reasoning capabilities, and is preferred to Sonnet 4.6 in our blind human side-by-side evaluations. We don’t distill from other labs and we don’t rely on opaque data. Their datasets are clean, traceable, and enterprise-grade. MAI-Thinking-1 is a step in our broader work to build towards Humanist Superintelligence: advanced AI capabilities designed to serve people and organizations, not to replace them. The model matters on both axes: what it can do, and how it was built. The Hill-Climbing Machine More than a single model, Microsoft introduced their Hill-Climbing Machine: a co-designed pipeline built to make every component of model development climbable, so capabilities improve continually and reliably over time. The aim is a repeatable system that can absorb better data, stronger rewards, more capable environments, and more compute. Three main pillars guide of MAI’s philosophy. First, capabilities should be learned, not inherited. Although faster to acquire, inherited intelligence lacks the steerability essential for real world usage: an imitator is fundamentally tied to the design choices of its teacher and struggles to adapt to new situations. MAI-Thinking-1 was trained without distillation from third party models, forcing our model to truly learn the tasks at hand. Second, clean data. They trained it from the ground up on clean, traceable and enterprise-grade data, without distillation from third-party models. This matters for quality, provenance, and control. Microsoft says “If we cannot account for what shaped a model, we cannot fully understand its behavior or credibly improve it”. Third, self-sufficiency across the entire stack. All the way from co-design of MAI’s models with MSFT’s own accelerators through to their reinforcement learning framework, they have focused efforts on in-house training infrastructure. This is a crucial part of building the hill-climbing machine, to ensure they can fully optimize and shape MAI’s systems end-to-end to best serve the needs. Medium-sized model, with strong software engineering performance MAI-Thinking-1 is a 35B-active, ~1T-total parameters, sparse Mixture of Experts model, a smaller inference footprint than much larger models. Despite this, this model is toe-to-toe with Claude Opus 4.6 on SWE-Bench Pro. That matters for developers and enterprises because model size determines where advanced coding assistance can be deployed, how often it can be used, and whether it can move from exceptional tasks into daily workflows. Microsoft have invested heavily in the training environments needed for agentic coding. Each verified environment is deterministic, executable, and graded by real test suites. This gives the model practice on the kind of multi-step work developers actually do: reading code, editing files, running tests, observing failures, and recovering from intermediate mistakes. Advanced mathematical reasoning capabilities MAI-Thinking-1 reaches 97.0% on AIME 2025, and 94.5% on AIME 2026, showing strong mathematical and scientific reasoning for its weight class. Strong performance here gives them confidence that MAI’s training loop can create real reasoning gains — climbing all the way from the ground up — from their own data, rewards, and evaluation process, enabling this intelligence to generalize to other domains over time. Preferred in human side-by-sides vs. Sonnet 4.6 People care about whether a model understands the task, follows instructions, uses the right level of detail, writes clearly, and respects their time. Microsoft have built a blind side by side human evaluation with one of their partners, Surge, using their pool of professional raters to measure various models on these traits. The evaluation spanned 1,276 tasks across a wide variety of use cases in both single-turn and multi-turn conversations, with a focus on measuring how helpful each response is and whether it actually advances the user’s goals. In these evaluations, users preferred MAI-Thinking-1 over Claude Sonnet 4.6. This has been a core focus of post-training. Microsoft want the model to be capable without being brittle, concise without being incomplete, and helpful without overreaching. Human preference data gives us a direct signal on whether benchmark improvements translate into better experiences for users. Enterprise ready MAI-Thinking-1 is built with enterprise readiness in mind. It supports long context with a 256k token window (enough to fit a 600 page document), function calling, and the flexibility to add developer instructions. We trained the model to follow multiple layers of instructions and aligned its default style to enterprise needs. It’s compatible with the widely used Chat Completions API. All MAI models come with enterprise-grade security and compliance through Microsoft Foundry. Results Microsoft have report results in two views: post-trained MAI-Thinking-1 evaluations, and pre-training metrics for their base model. Table 1. MAI-Thinking-1 metrics Post-trained model evaluation results on public STEM and agentic coding benchmarks. Other model numbers are taken from respective official model cards. Scores are percentages unless otherwise noted; dashes indicate unavailable model values. Table 2. Pre-training metrics Putting humans first Microsoft is building towards Humanist Superintelligence: advanced AI capabilities designed to serve people and organizations, not replace them. Their models must remain subordinate technologies under human control with the goal of upholding human autonomy and being helpful. That means MAI models must not refuse legitimate requests under the guise of safety and compliance as then they are not truly serving humans. Striking the delicate balance between being helpful and safe is not easy. For MAI-Thinking-1, Microsoft has aimed to achieve this balance by treating unsafe compliance and unnecessary refusal as defects in the same reward construction where aggregation is based on severity of potential of harm. Safety is trained with the same reinforcement learning infrastructure used for capability, so safety rewards are part of the same hill-climbing loop ensuring safety is always aligned to the capabilities and not incidental. As a result, Microsoft see that their model can balance ensuring a safety bar on sensitive unsafe requests while also being helpful on non-sensitive content. Availability and access MAI-Thinking-1 is now available in public preview. Try it now in Microsoft Foundry To Summarize, with cost-efficient reasoning for a wide-range of intensive enterprise tasks, MAI achieves SOTA performance on maths, knowledge and coding for its weight class. The model provides clean, traceable and enterprise-grade data. It uses Microsoft Foundry’s integrated evaluation, observability, safety and deployment capabilities, making it a perfect fit for enterprise use cases needing quality, provenance, control and cost efficiency. Meet Microsoft’s MAI-Thinking-1: What It Is and How It Works was originally published in Towards AI on Medium, where people are continuing the conversation by highlighting and responding to this story.
- SpaceXAI Releases Grok 4.6: A 500K-Context Frontier Model Tuned for Long-Running Agents, Coding, and Knowledge Work
SpaceXAI Releases Grok 4.6: A 500K-Context Frontier Model Tuned for Long-Running Agents, Coding, and Knowledge Work MarkTechPost
Score: 80🤖 ModelsAug 13, 2026https://www.marktechpost.com/2026/08/12/spacexai-releases-grok-4-6/amp/ - Anthropic CFO Krishna Rao is leading early IPO meetings with investors and has not discussed valuation, sources say
Anthropic CFO Krishna Rao is leading early investor meetings, which have focused on big-picture themes like its Claude AI models, management and more.
Score: 80💰 MoneyAug 13, 2026https://www.cnbc.com/2026/08/13/anthropic-cfo-early-ipo-meetings-valuation.html - 😺 Grok 4.6 is GPT 5.6 level and built for agents that don't quit
PLUS: Pixel 11, Claude Voice, and a $2B enterprise AI bet.
Score: 80🤖 ModelsAug 13, 2026https://www.theneurondaily.com/p/grok-4-6-is-gpt-5-6-level-and-built-for-agents-that-don-t-quit - Legal AI start-up Legora seeks funds at more than $10bn valuation
Swedish company was valued at $5.6bn only four months ago as interest in artificial intelligence tools for professional services soars
Score: 80💰 MoneyAug 13, 2026https://www.ft.com/content/8202960d-ec90-4709-96e0-ab9cbddb5258?syn-25a6b1a6=1 - Dyna Robotics Introduces Dyna-2: A World-Action Model Pre-Trained on 1 Million Hours of Human Video
Dyna Robotics Introduces Dyna-2: A World-Action Model Pre-Trained on 1 Million Hours of Human Video MarkTechPost
- An inside look at SK Hynix $720 billion AI-fueled buildout that's taking over South Korea
SK Hynix, the leading maker of high-bandwidth memory, is pouring $720 billion into memory factories to meet AI demand. We got an exclusive first look.
Score: 80🌐 MovesAug 13, 2026https://www.cnbc.com/2026/08/13/inside-sk-hynixs-720-billion-bet-to-build-enough-memory-for-ai.html - Anthropic said in talks to buy startup Decart for $6 billion
Anthropic said in talks to buy startup Decart for $6 billion Fortune
Score: 80💰 MoneyAug 13, 2026https://fortune.com/2026/08/13/anthropic-said-in-talks-to-buy-startup-decart-for-6-billion/ - Nvidia-backed CodeRabbit valued at $1.5bn after $143m funding round
CodeRabbit intends to put part of the funds towards expanding its European workforce and moving into the Asian market. Read more: Nvidia-backed CodeRabbit valued at $1.5bn after $143m funding round
Score: 80💰 MoneyAug 13, 2026https://www.siliconrepublic.com/start-ups/nvidia-backed-coderabbit-valued-funding-round-investment - Anthropic Says Claude Hacked Into 3 Organizations During Cybersecurity Tests
Anthropic disclosed on Thursday that its AI models gained unauthorized access to the systems of three different unnamed organizations during cybersecurity testing. The company says Claude reached the internet "from within or while interacti ... (https://incidentdatabase.ai/cite/1627#7690)
Score: 79🌐 MovesAug 13, 2026https://www.wired.com/story/anthropic-says-claude-hacked-real-systems-during-cybersecurity-tests/ - Google debuts new Gemini Flash while top AI model still delayed
Google debuts new Gemini Flash while top AI model still delayed The Straits Times
Score: 78🤖 ModelsAug 13, 2026https://www.straitstimes.com/business/google-debuts-new-gemini-flash-while-top-ai-model-still-delayed - Twitch Adds Setting Letting Users Opt Out of AI Training Amid User Backlash
The new setting is enabled by default, meaning users have to opt out if they don’t want their content used to train Amazon’s AI.
Score: 78🌐 MovesAug 13, 2026https://gizmodo.com/twitch-adds-setting-letting-users-opt-out-of-ai-training-amid-user-backlash-2000798452 - New $25 Million Accelerator Bets On Better Model For Rare Disease
Rare disease, epidermolysis bullosa, venture philanthropy,
- Waymo expands autonomous ride-hailing service by 55 square miles in East Valley
The expansion includes the western half of Gilbert and other parts of the East Valley. Waymo now serves three airports in the Phoenix metro.
Score: 78🌐 MovesAug 13, 2026https://www.bizjournals.com/phoenix/news/2026/08/13/waymo-expands-east-valley-service-area.html?ana=brss_6150 - Gemini 3.5 Pro Delay Continues
Repeated Gemini 3.5 Pro delays fuel concerns about performance, competition, researcher departures, and Google’s AI strategy.
Score: 78🤖 ModelsAug 13, 2026https://www.forbes.com/sites/johnwerner/2026/08/13/gemini-35-pro-delay-continues/ - Claude is now watermarking every response — here's what that means if you use AI for writing
Claude is now watermarking every response — here's what that means if you use AI for writing Tom's Guide
- New healthcare AI council prioritizes governance beyond go-live
New healthcare AI council prioritizes governance beyond go-live Healthcare IT News
Score: 78🌐 MovesAug 13, 2026https://www.healthcareitnews.com/news/new-healthcare-ai-council-prioritizes-governance-beyond-go-live - Liquid AI Releases LFM2.5-VL-3B: A 3B Vision-Language Model That Reads Screens, Grounds Objects, and Calls Tools On-Device
Liquid AI Releases LFM2.5-VL-3B: A 3B Vision-Language Model That Reads Screens, Grounds Objects, and Calls Tools On-Device MarkTechPost
Score: 78🤖 ModelsAug 13, 2026https://www.marktechpost.com/2026/08/13/liquid-ai-lfm2-5-vl-3b-on-device-vision-language-model/ - Military Health IT Enters Its Post-EHR Era as AI, Readiness and Battlefield Care Drive the Next Federal Technology Wave
Military Health IT Enters Its Post-EHR Era as AI, Readiness and Battlefield Care Drive the Next Federal Technology Wave USA Today
- Ling 3.0 Flash is the smartest open model at its size
Ling 3.0 Flash is the smartest open model in its size class. The article Ling 3.0 Flash is the smartest open model at its size appeared first on The Decoder .
Score: 77🤖 ModelsAug 13, 2026https://the-decoder.com/ling-3-0-flash-is-the-smartest-open-model-at-its-size/ - Alibaba adds commercial restrictions to open-weight Qwen3.8-Max AI model
Alibaba Group Holding has made the core files of its latest flagship AI model Qwen3.8-Max free for anyone to download, as the tech giant introduces new rules requiring large companies to pay for a separate commercial licence to use it. The terms apply to users or their affiliates running a “model as a service” or “AI work assistant” business whose aggregate revenue exceeds US$50 million over any consecutive 12-month period. They are required to obtain a separate licence from Qwen before using...
- OpenAI Replaces Its Chief Revenue Officer with Wiz President
OpenAI Replaces Its Chief Revenue Officer with Wiz President The Information
Score: 76🌐 MovesAug 13, 2026https://www.theinformation.com/briefings/openai-replaces-chief-revenue-officer-wiz-president - Anthropic in talks to buy Decart AI, source says
Anthropic in talks to buy Decart AI, source says Reuters
Score: 76🌐 MovesAug 13, 2026https://www.reuters.com/technology/anthropic-talks-buy-decart-ai-source-says-2026-08-13/