Edition: June 19, 2026

The US government has ordered Anthropic to restrict the export of its powerful AI models Fable and Mythos, citing national security concerns, in a move that may set a precedent for the regulation of AI technology. The episode highlights the challenges of using export controls to contain the spread of advanced technologies. The effectiveness of such controls has been uneven in the past, as seen in the cases of encryption and spyware.

Billionaire Mukesh Ambani's Reliance Industries is positioning itself as a leader in India's AI race, unveiling new AI services for phone calls, mobile apps, and connected homes. The company aims to embed AI assistance directly into its telecom network, reducing reliance on third-party apps. Reliance's AI ambitions include investing $110 billion in AI infrastructure to establish itself as a major player in India's emerging AI ecosystem.

SAP and Google Cloud have partnered to deploy an agentic commerce architecture, aiming to automate multi-agent marketing and retail operations at an enterprise scale. This move addresses the structural data failure in customer experience and CRM platforms, and enables a standardized data exchange among retailers, payment gateways, and autonomous agents. The partnership seeks to improve customer experience and retention through AI-driven commerce operations.

A startup claims to have broken through a mathematical bottleneck holding back large language models, while brain-computer interface trials are gaining momentum with advances in technology and increasing numbers of volunteers. These developments could lead to faster, cheaper, and more accessible AI and BCI technologies.

Miami-based AI startup Subquadratic claims to have broken through a mathematical bottleneck holding back large language models, developing a new model called SubQ that is faster, cheaper, and more energy-efficient. Independent evaluation results suggest the company's claims may be valid, potentially changing how LLMs are built. SubQ could offer huge increases in speed at a fraction of the typical cost for certain tasks.

e2e-assure introduces Cumulo, the UK's first sovereign AI-driven zero-day SOC platform, designed to secure IT and OT environments with early threat identification and vulnerability detection. Cumulo utilizes digital twin technology and customer-dedicated AI models to provide a proactive defense against AI-driven threats. The platform aims to revolutionize traditional SOC models with its AI-first approach and human oversight.

Brain-computer interface trials are gaining momentum, with a growing number of volunteers and advancements in technology enabling people with paralysis to communicate and regain independence. One such example is Casey Harrell, who has been using a BCI to 'speak' and perform daily tasks despite being paralyzed with ALS. The number of BCI trial volunteers has soared in recent years, with China becoming the first country to approve a BCI for medical use.

The US Commerce Department has expressed concerns that ASML's extreme ultraviolet lithography machine, a crucial tool for advanced semiconductor production, may have been shipped to China, despite ASML's denial of the claim. The alleged breach of export controls could have significant implications for the global AI buildout and the US-China tech rivalry. ASML is a key player in the chip industry, with its machines being the only ones capable of printing the most advanced semiconductor patterns.

Elastic has agreed to buy DeductiveAI, a startup using AI to catch and resolve software bugs, for up to $85 million. The acquisition reflects a trend of established tech companies buying AI-native startups to integrate their technologies. The deal is expected to enhance Elastic's observability platform with Deductive's AI technology.

The US government has forced Anthropic to pull its two newest models due to national security concerns, but cybersecurity researchers argue the move is unnecessary and may even benefit the company. The ban has sparked debate over its implications for developers and the company's IPO. The move may inadvertently boost Anthropic's brand.

The US government has banned Anthropic's Fable 5 release due to national security concerns, but experts question the move's validity. The ban may have unintended benefits for the company. The incident highlights the complex relationship between Anthropic and the US government.

Amazon MGM has dropped the film 'Artificial' about OpenAI CEO Sam Altman, which was directed by Luca Guadagnino and starred Andrew Garfield. The film covered Altman's termination and reinstatement as CEO in 2023. The studio believes the movie will be better served by a different studio.

Allbirds, a former shoe company, has pivoted to AI and rebranded as Smartbird, with new CEO Nadia Carlsten leading the charge to provide AI infrastructure services, targeting customers who require direct control over servers and value data sovereignty. The company aims to deploy compute clusters for several customers by the end of the year. Smartbird's success is uncertain, but it has secured $100 million in funding and is recruiting a new team.

The author reflects on their decade-long experience with self-quantification, tracking personal data to gain self-knowledge, but ultimately realizing the limitations and potential corruption of metrics. The pursuit of numerical goals led to an obsession with step counts, rather than the initial goals of physical and emotional well-being. The article highlights the duality of measurement, revealing both useful insights and obscured truths.

Barret Zoph, OpenAI's head of enterprise AI sales, has left the company after just five months, following his return in January. Zoph's departure comes as OpenAI focuses on key revenue drivers ahead of its planned IPO. His role was significant in leading the company's push into enterprise sales.