Baseten, a company providing inference software for running AI models on GPUs, is reportedly in discussions to raise $1 billion at an $11 billion valuation, more than doubling its valuation in just three months. This highlights strong investor confidence and rapid growth in AI inference infrastructure.
Sources: Latent Space, The Neuron
The US government is tracking 'anti-technology extremists' amid protests against AI data centers and job fears, while the Catholic Church warns against the ethical implications of AI's rapid expansion. This highlights data centers as a critical political fault line where economic growth and societal concerns converge.
Sources: The Neuron
Nvidia CEO Jensen Huang has reportedly identified a "brand new" market worth $200 billion for the company. This indicates Nvidia's continued expansion and strategic focus on new opportunities within the technology sector, likely related to AI given Nvidia's core business.
Sources: Hacker News
Corey Quinn highlighted Anthropic co-founder Christopher Olah's significant influence on the 'Magnifica Humanitas' document, suggesting it canonized specific technical limitations of an AI product as a spiritual treatise, an unprecedented act of vendor lobbying in AI ethics.
Sources: Simon Willison
OpenRouter raised $113 million from CapitalG and other investors, following a surge in weekly volume to 25 trillion tokens across 400+ models and 8 million users. This significant funding highlights the growing importance and demand for AI model routing services.
Sources: The Neuron
A new research paper, "Self-Verified Distillation," proposes an algorithm enabling large language models (LLMs) to improve themselves using only unlabeled prompts and self-verification, without external teachers. This method, applied to Qwen3 models, achieved significant performance gains in math, science, and coding by allowing LLMs to generate, filter, and train on their own curated datasets.
Sources: arXiv AI
Anthropic is significantly expanding the enterprise adoption of its AI model, Claude, through a global strategic alliance with KPMG, integrating it across their 276,000-strong workforce. Concurrently, Anthropic is strengthening its presence in the high-growth Korean market by appointing a new director and preparing to open a Seoul office, while also securing partnerships with SK Telecom for customer service and Law&Company for legal AI.
Sources: Anthropic News
Amazon Bedrock has introduced AgentCore Payments, a new managed service in preview that enables generative AI agents to autonomously handle microtransactions for paid APIs, content, and managed compute platforms. This service addresses key challenges like secure agent funding, fragmented payment protocols, real-time spending limits, and comprehensive observability, aiming to accelerate the development of agentic payment workflows.
Sources: AWS ML Blog
Elon Musk announced that xAI has finished training Grok V9-Medium, a large 1.5 trillion-parameter model, with an expected release in 2-3 weeks. This marks a significant development for xAI's competitive offering in the large language model space.
Sources: The Neuron
Qualcomm has secured a deal to sell millions of AI chips to ByteDance for its data centers, indicating a major partnership in AI infrastructure. This strengthens Qualcomm's position in the AI chip market and supports ByteDance's expanding AI capabilities.
Sources: The Neuron
Anthropic's revenue grew fivefold in the first five months of 2026, reaching an annualized rate of approximately $45 billion, surpassing OpenAI's reported revenue. This demonstrates Anthropic's rapid commercial success and strong competitive position in the AI market.
Sources: The Neuron
Figure AI has secured its first large-scale retail supply chain deal, deploying its Figure 03 humanoids at JCPenney's distribution center in Reno, Nevada. This marks a significant step for humanoid robots entering practical commercial applications in logistics.
Sources: The Neuron
The Verge investigated how AI weapons, including Project Maven, and military procurement are accelerating the transition of AI warfare from theoretical concepts to practical deployment. This raises significant ethical and geopolitical concerns about the future of conflict.
Sources: The Neuron
DeepSWE is a new benchmark designed to evaluate AI coding agents on complex engineering tasks, featuring prompts that are half the length of SWE-bench Pro. This release is significant for advancing the evaluation and development of more capable AI software engineering tools.
Sources: Digg AI
Yo Shavit, OpenAI's frontier AI safety policy lead, has resigned from his position to join the OpenAI Foundation's AI Resilience program. He cited a limited timeline to prepare for superintelligence, indicating a continued focus on long-term AI safety and governance within the OpenAI ecosystem.
Sources: Digg AI