King Charles III met with leaders from OpenAI, Anthropic, Google DeepMind and Nvidia at Dumfries House in Cumnock, Scotland on September 17, 2026. The king urged AI developers to keep the technology under control and ensure it serves society in ways that are transparent and accountable. Experts have warned that AI could pose a significant threat to humanity if not developed responsibly.
ASUS and Poesis completed a week-long experiment showing that autonomous AI agents can trade stocks effectively using local hardware. The test ran on an ASUS ExpertCenter Pro ET900N G3 laptop powered by NVIDIA technology, with eight AI portfolio-manager agents running different strategies on a single machine. ASUS estimates local deployment costs about $4,000 upfront plus $384 yearly in electricity, compared to $6,000 to $8,000 in annual cloud API fees.
Kimi K3, an open-weight model from Moonshot AI, is now available on Amazon Bedrock. The model supports tool calling, structured output, reasoning and response streaming, with zero data retention during inference. Users can deploy it through a global cross-Region profile that costs roughly 10 percent less than geographic profiles.
AI-linked PACs have spent nearly $1 million in a South Dakota Senate race, covering advertising, mailers and consultant fees. The spending raises fresh questions about how AI industry money influences politics and whether campaign finance laws keep pace with this kind of political engagement.
Key Takeaways
- King Charles III met with OpenAI, Anthropic, Google DeepMind and Nvidia leaders in Scotland to urge responsible AI development
- ASUS and Poesis demonstrated AI stock trading on local hardware, saving an estimated $5,600 to $7,600 versus annual cloud costs
- Eight AI portfolio-manager agents ran investment strategies on a single ASUS machine powered by NVIDIA GB300 hardware
- Moonshot AI launched Kimi K3 on Amazon Bedrock with zero data retention and lower cross-Region pricing
- AI-linked PACs spent nearly $1 million in a South Dakota Senate race, raising transparency concerns
- Apollo Global Management is investing in AI hardware startups including chip companies and data center operators
- Kornic Automation will invest 8 billion won in an AI and software R&D center in Daejeon, South Korea
- A new malware called RedHat uses AI to autonomously control Android devices and reinstall deleted components
- Researchers developed a traffic rule translation framework achieving 75 percent accuracy, outperforming baseline methods at 60 and 55 percent
- Fifty-two percent of financial firms have experimented with agentic AI according to Cambridge Judge Business School
King Charles meets AI leaders to discuss safety concerns
King Charles III met with leaders from OpenAI, Anthropic, Google DeepMind and Nvidia in Scotland. He urged them to keep artificial intelligence under control and beneficial for people. The meeting took place at Dumfries House in Cumnock, Scotland. The king wants AI development to be transparent and accountable.
King Charles urges AI leaders to keep technology under control
King Charles III urged senior AI leaders to make sure artificial intelligence remains under control and serves society. The meeting was held on Thursday, September 17, 2026. The king met with industry leaders, government officials and thought leaders. Safety concerns about AI are growing among experts.
King Charles warns AI must serve society at Scotland meeting
King Charles III met with leaders from OpenAI, Anthropic, Google DeepMind and Nvidia at Dumfries House in Cumnock, Scotland. He stressed that AI should be used for the greater good in ways that are transparent and accountable. Experts warn AI could pose a significant threat to humanity if not developed responsibly. The king promotes a better understanding of AI benefits and risks.
ASUS and Poesis test AI agents that trade stocks locally
ASUS and Poesis completed a week-long experiment where autonomous AI agents conducted market research and executed real trades. The test ran on an ASUS ExpertCenter Pro ET900N G3 laptop with NVIDIA technology. ASUS argues local deployment saves money compared to cloud costs, estimating about $4,000 upfront plus $384 yearly in electricity versus $6,000 to $8,000 in annual cloud API fees. Fifty-two percent of financial firms have experimented with agentic AI according to Cambridge Judge Business School.
ASUS and Poesis run autonomous AI trading on local supercomputer
ASUS and Poesis ran a week-long experiment using the ASUS ExpertCenter Pro ET900N G3 deskside supercomputer powered by the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip. The system delivers up to 20 PFLOPS of AI performance with 748GB of coherent memory. Eight AI portfolio-manager agents ran different investment strategies on a single machine using NVIDIA Nemotron 3 Super. The agents handled research, risk management and trade execution using live capital without relying on cloud infrastructure.
Apollo Invests in AI Hardware Startups for Infrastructure Financing
Apollo Global Management is investing in AI hardware startups like chip companies and data center operators. The company plans to offer equity and debt financing to support AI infrastructure projects. Apollo has about 60 employees focused on digital infrastructure deals. The strategy includes backing companies like Hadrian and SiFive that raised billions in recent funding rounds.
Agentic AI Networking for Drones in Low-Altitude Wireless Networks
Researchers proposed a new AI system to manage drones and other flying devices in low-altitude wireless networks. The system uses large language models and reinforcement learning to coordinate multiple services sharing airspace. It adapts to changing conditions without needing to retrain. A logistics monitoring case study showed the framework can handle complex coordination among different services.
New Method Translates Traffic Rules into Computer Logic More Accurately
A new translation framework called S4L to Prolog converts traffic rules written in plain language into executable computer logic. It achieved 75 percent accuracy on real-world traffic rules, outperforming two baseline methods that scored 60 and 55 percent. The framework handles implicit rules, exceptions, and causal relationships better. The study used a benchmark of twenty real traffic rules to test each approach.
Kimi K3 AI Model Now Available on Amazon Bedrock
Moonshot AI launched Kimi K3, an open-weight model, on Amazon Bedrock. The model supports tool calling, structured output, reasoning, and response streaming. Users can deploy Kimi K3 with zero data retention and no operator access during inference. It can be called through a global cross-Region profile that costs about 10 percent less than geographic profiles.
Probabilistic Allen Algebra Extends Interval Relations for Uncertainty
Researchers developed a probabilistic extension of Allen's interval algebra to handle uncertain temporal information. Allen's original system uses thirteen crisp relations based on exact interval boundaries, which struggles with imprecise data. The new Probabilistic Allen Algebra addresses uncertainty from language, perception, and databases. The work aims to improve reasoning about time in situations where exact boundaries are unknown.
Kornic Automation to Invest 8 Billion Won in AI and SW R&D Center in Daejeon
Kornic Automation, an affiliate of APS Group, will invest 8 billion won to build an AI and software R&D center in Yuseong-gu, Daejeon. The center will expand the company's EasyCluster equipment-control software into robotics and factory automation. It will combine AI with EasyCluster to create automation solutions for semiconductor and secondary-battery manufacturing. The facility will also serve as a hub for physical AI research in the Daedeok Techno Valley area.
AI PACs Have Dumped Nearly $1 Million Into an Obscure Senate Race
PACs linked to AI labs and investors have spent nearly $1 million in a South Dakota Senate race. The spending covered advertising, mailers, staff, and consultants. The PACs are supporting the incumbent and opposing the challenger in the reliably Republican seat. The spending raises questions about the influence of AI money in politics and the transparency of campaign finance laws.
New Android malware can deploy AI to automate device control and bring itself back from the dead
Zimperium zLabs discovered RedHat, a Chinese-origin Android banking trojan that uses AI to control devices. The malware spreads through third-party app stores, social media, malvertising, and SMS spam. It uses an AI assistant to navigate the device interface in real time, making it more adaptable than traditional malware. RedHat can also reinstall deleted components and intercept uninstall attempts to persist on a device.
Sources
- The king and AI: UK monarch Charles meets artificial intelligence leaders
- The king and AI: UK monarch Charles meets artificial intelligence leaders as safety concerns swirl
- The king and AI: UK monarch Charles meets artificial intelligence leaders as safety concerns swirl
- ASUS and Poesis Test AI Agents That Research and Trade, Sparking Local Deployment Debate
- ASUS and Poesis test autonomous AI trading on local system
- Apollo Backs AI Hardware Startups to Win Future Infrastructure Financing
- Agentic AI Networking for Heterogeneous Unmanned Aerial Systems in Low-Altitude Wireless Networks
- Structured Four-Stage Legal Translation: From Natural-Language Traffic Rules to PROLOG
- Introducing Kimi K3 on Amazon Bedrock
- PAA: The Probabilistic Allen Algebra: A Generative and Complete Probabilistic Extension of Allen's Interval Relations
- Kornic Automation to Invest 8 Billion Won in AI and SW R&D Center in Daejeon
- AI PACs Have Dumped Nearly $1 Million Into an Obscure Senate Race
- New Android malware can deploy AI to automate device control — and it can even bring itself back from the dead
Comments
Please log in to post a comment.