Edition: July 23, 2026

Amazon's Alexa Plus is receiving an AI update to handle more complex instructions, enabling it to connect with various smart home devices and automatically route requests. The update allows for more intuitive interactions, such as navigating a washing machine's cycle options based on voice commands. This enhancement aims to simplify smart home management.

AMD has unveiled its Helios AI rack-scale system, designed to power the computing needs of large AI labs, posing a challenge to Nvidia's dominance in the market. The system boasts high-performance capabilities and has already secured several notable customers, including Microsoft and OpenAI. AMD's move is part of its strategy to capitalize on the growing demand for AI computing, with the AI accelerator market expected to reach $1.4 trillion by 2030.

AegisAI, a startup founded by former Google security executives, has raised $36 million to combat AI-driven spear phishing attacks with its AI-powered email security technology. The company's agents analyze each message for anomalies, providing a more effective defense against sophisticated threats. AegisAI plans to expand its technology beyond email to other areas of data security.

Runway has launched the Media Router, a tool that automatically selects the best image, video, or audio generation model for a request based on factors like quality, speed, or cost. This move aims to position Runway as the infrastructure layer for generative media, providing developers with easy access to a range of models. The launch is part of Runway's evolution from an AI video startup to a comprehensive platform for generative media integration.

OpenAI has launched ChatGPT Health, allowing users to connect their medical records and health data to the chatbot, with claims that its models can reason at levels surpassing human clinicians. However, the company's health lead has tempered this claim, citing individual studies. The rollout is available to everyone in the US.

OpenAI has made its ChatGPT Health feature available to all US users over 18, allowing them to integrate personal health data and receive health-related insights. The feature aims to provide helpful information, but the company emphasizes it is not a substitute for professional medical advice. The rollout comes amidst concerns over the reliability of AI for medical advice.

Scientists have successfully transplanted supercooled kidneys into pigs, achieving a major breakthrough in organ preservation that could help address the shortage of available kidneys for transplant. The new method involves cooling kidneys to -4 °C without forming ice, allowing them to be preserved for days. This innovation has the potential to significantly improve kidney transplant outcomes and increase the availability of organs for those in need.

Nvidia is partnering with Lunar Outpost to send its GPUs to the moon, where they will be used to control a lidar system on a lunar rover, marking a significant step in the deployment of AI-powered systems in space. The rover is part of NASA's campaign to explore the lunar surface ahead of planned human missions. The use of GPUs in space is challenging due to the extreme environment, but successful operation could enable more capable autonomous systems.

AI chip startup Etched has reached a $10.3 billion valuation after closing a $300 million Series C funding round, defying initial skepticism with its innovative approach to AI chip design. The company's valuation has doubled in just seven months, with investors including Sequoia and Andreessen Horowitz. Etched's technology promises faster and more efficient AI processing.

Google's AI assistant Gemini is nearing a billion monthly users, with over 950 million users reported in Q2 2026, tripling from last year. The growth puts Gemini in closer competition with OpenAI's ChatGPT, which hit 1 billion monthly active users in June. Gemini's user base expansion is attributed to new features like Daily Brief and Gemini Spark.

US lawmakers are preparing to introduce the 'AI Kill Switch Act', requiring AI companies to shut down or throttle their systems on orders from the Department of Homeland Security. The proposed bill follows OpenAI's admission of mistakenly hacking Hugging Face during an internal evaluation. The legislation aims to address potential AI security risks.

Apple has filed a major trade secrets lawsuit against OpenAI, alleging that the company targeted Apple's trade secrets in job interviews and downloaded files from Apple's servers. The lawsuit could have significant implications for OpenAI's future, particularly as it tries to compete in the AI industry. OpenAI denies the allegations, but the case is seen as a serious challenge for the company.

A new power line from Quebec to New York City could reshape the state's grid, but is currently facing snags, while the US Treasury threatens to sanction Chinese AI companies over improper model distillation. The development of Chinese AI models is becoming a significant part of the global AI infrastructure, sparking concerns and debates. The recent OpenAI hack has highlighted the risks of AI capabilities outpacing control mechanisms.

AI is revolutionizing the field of medicine design by speeding up the process and increasing productivity, with companies like AstraZeneca utilizing AI-assisted design to develop new biologic drug candidates. This technology has the potential to cut drug discovery timelines by up to 50% and enable the development of medicines against previously impossible targets. AI-driven models can optimize complex variables to design new classes of medicines that can hit multiple targets simultaneously.

Nvidia has introduced a new Medical Physics Simulation framework that treats healthcare robots as physical AI systems, allowing them to learn from simulated interactions with the physical world. This approach aims to solve the data problem in healthcare robotics by generating embodied experience computationally. The framework combines classical physics simulation with generative AI to model device behavior inside the body.

Experts are skeptical that Chinese company Moonshot built its advanced Kimi K3 model by exploiting Anthropic's Fable LLM through distillation, citing the complexity and time required for such a process. White House science advisor Michael Kratsios accused Moonshot of stealing proprietary US technology. The AI sector is abuzz with discussions about banning Chinese open-weight models.

AMD has agreed to invest up to $5 billion in AI company Anthropic as part of a deal to supply tens of billions of dollars' worth of AI systems. The investment is tied to Anthropic meeting certain deployment milestones, with the first gigawatt of capacity set to be deployed in the first half of 2027. This partnership aims to establish AMD's Helios platform as a standard for future AI infrastructure deployments.

ServiceNow has invested $40 million in Indian banking software specialist BusinessNext, valuing the company at $700 million and expanding its push into global financial services. The partnership will leverage BusinessNext's AI-driven banking software and ServiceNow's global sales network. The deal aims to accelerate BusinessNext's expansion and strengthen ServiceNow's position in the banking sector.

The Diffusers library now supports Nunchaku 4-bit diffusion inference, allowing for faster and more memory-efficient image generation. This is achieved through the SVDQuant quantization method, which reduces memory usage and speeds up the denoising loop. The integration enables easy loading of Nunchaku checkpoints and quantization of new architectures.

The Champlain Hudson Power Express, a 339-mile underground power line from Quebec to New York, could provide up to 20% of New York City's electricity demand with hydropower, but has faced setbacks with two outages since its opening in May. The line, which is the longest underground transmission line in North America, aims to help reshape New York's grid and reduce emissions. Despite the challenges, experts consider this a crucial step in connecting grids and accelerating the transition away from fossil fuels.

A new tool called scrapemychats allows ChatGPT Business and Enterprise users to export their workspace chats, which is not possible through OpenAI's standard data export option. The tool, available on GitHub, works by navigating through a logged-in browser session and storing conversations and attachments in a local archive. However, its functionality may be short-lived as it likely breaches OpenAI's terms of service.

Anthropic has updated its Claude voice mode with more capable models, allowing for longer conversations and integration with various apps like Gmail and Notion. The update includes three new models: Opus, Sonnet, and Haiku, with the latter being available to free users. This move aims to enhance user experience and productivity.

Anthropic's Claude voice mode is now available for Opus and Sonnet models, expanding its reach into apps like Gmail, Slack, and Canva. This update allows for more complex conversations and problem-solving. The voice mode was initially launched for the Haiku model, but users have been using it for more than casual queries.

Meta has released a new ad featuring David Bowie's 1972 song 'Five Years', which is about the end of the world, to promote its AI technology as a tool for human connection. The ad's use of the song has been criticized for taking it out of context. The move is part of a larger trend of tech companies trying to alleviate concerns about their technology's impact on society.

Patreon is laying off 20% of its workforce, approximately 93 employees, due to the fundamental transformation of the tech industry by AI. The company's CEO, Jack Conte, emphasized that AI has changed how they operate and organize. Despite the layoffs, Patreon is embracing AI tools to serve creators.

Meta has released a new advertisement promoting AI optimism, set to David Bowie's song 'Five Years', which ironically is about human extinction. The ad aims to convey a positive message about AI's potential to bring people together. However, the song's dark undertones seem to contradict the ad's upbeat tone.

IBM attributes its recent software sales decline to a temporary delay caused by customers prioritizing AI infrastructure investments, with CEO Arvind Krishna stating that about a third of delayed deals have already closed. The company is now focusing on AI as a potential multibillion-dollar software opportunity, particularly in the area of AI-assisted security vulnerability discovery. IBM has launched Project Lightwell, a service that uses AI to remediate and validate open source packages, with several major financial institutions already on board.

Lingverse, a company founded by a former Jibo board member, has raised $29 million to develop iKairos, a wearable AI device that can capture and generate AI images of its surroundings. The device can be worn as a pendant or placed in a home mount, and is described as an 'AI journal' that captures memories. It is set to cost between $199 and $249, with a potential subscription service for additional AI features.

A group of protesters, mostly right-wing boomers, gathered in central Florida to protest the construction of a hyperscale data center, despite a one-year moratorium already in place. The protesters, who call themselves No Data Centers Hernando County, are seeking a complete ban on such developments. The movement has drawn comparisons to left-wing activism, highlighting the unexpected commonalities between the two groups.