US Agencies Warn of AI-Generated Code Hacking Critical Infrastructure

US federal agencies have warned of an active threat where attackers are using AI-generated code to hack into critical infrastructure controllers, targeting Siemens S7 Series programmable logic controllers (PLCs) at water, manufacturing, energy, and other critical facilities. The attackers use open-source industrial automation libraries and AI to create custom tools that mimic operational technology (OT) monitoring software.

Microsoft is investing heavily in AI and data centers, committing tens of billions of dollars to support large language models and generative AI tools. The company's Azure AI and Azure OpenAI Service power AI workloads, and it is building custom Maia AI chips and data centers.

Broadcom is in talks with lenders to raise over $60 billion in debt for an AI chip financing deal that will benefit Anthropic and other companies. The deal could include a $30 billion junior debt tranche and is part of Broadcom's efforts to finance its AI chip business.

OpenAI is slowing down training on some of its most advanced AI models to strengthen safety measures. The company will expand systems to detect dangerous behavior and add safety checks before restarting larger-scale training.

Google has a patent for 'Contextual estimation of link information gain,' which aims to reward content that offers something genuinely new. The patent proposes a way to score documents based on the additional information they contribute beyond what a user has already seen within a given search session.

Val Kilmer, who had a tracheotomy that affected his vocal cords, was brought back to life using AI technology in a posthumous film. The production team used old images and recordings of Kilmer's voice to create an AI-generated version of the actor.

Scale AI and other companies are expected to benefit from Broadcom's AI chip financing deal. Meanwhile, Morgan Stanley warns that AI data centers face a 38-gigawatt power gap, which could grow to 50 gigawatts by 2025.

Fortinet has acquired Virtue AI to expand its AI security capabilities. The deal adds Virtue AI's Guardian Agent capabilities to Fortinet's Security Fabric, enabling organizations to protect AI systems from development through live operation.

No-Filter 'Kriminal' AI Platform Raises Cybercrime Concerns, as it offers guardrail-free social engineering, offensive cybercrime, and OSINT scanning.

Key Takeaways

• US federal agencies warn of AI-generated code hacking critical infrastructure controllers, targeting Siemens S7 Series PLCs. • Microsoft invests heavily in AI and data centers, committing tens of billions of dollars. • Broadcom seeks $60 billion in AI debt deal, benefiting Anthropic and other companies. • OpenAI slows AI training over safety concerns, expanding systems to detect dangerous behavior. • Google patents 'Contextual estimation of link information gain' to reward new content. • Val Kilmer brought back to life with AI in new film. • Morgan Stanley warns of 38-gigawatt power gap for AI data centers. • Fortinet acquires Virtue AI to expand AI security capabilities. • No-Filter 'Kriminal' AI platform raises cybercrime concerns. • BigBear.ai adds defense expert to board.

Feds Warn of AI-Made Code Hacking Critical Infrastructure

US federal agencies have warned of an active threat where attackers are using AI-generated code to hack into critical infrastructure controllers. The attackers are targeting Siemens S7 Series programmable logic controllers (PLCs) at water, manufacturing, energy, and other critical facilities. The attackers use open-source industrial automation libraries and AI to create custom tools that mimic operational technology (OT) monitoring software. The feds urge critical infrastructure owners and operators to inventory PLCs, apply security patches, and ensure no PLCs are accessible from the internet.

AI-Generated Scripts Target Siemens PLCs in US Critical Infrastructure

The US government has warned of an ongoing activity targeting Siemens S7 Series PLCs to conduct reconnaissance and capability development using AI-generated scripts. The activity is broader in scope than Siemens PLCs and targets critical infrastructure sectors such as energy, water and wastewater, food and agriculture, and chemical. The attackers are using AI to generate exploitation scripts and rapidly iterate them, lowering the technical barriers to Industrial Control System (ICS) attacks.

Hackers Use AI to Target Siemens PLCs in Critical US Sectors

Several US government agencies have issued a joint cybersecurity advisory warning critical infrastructure organizations about hacker attacks targeting Siemens programmable logic controllers (PLCs). The hackers are scanning the internet to identify exposed PLCs and developing exploits that could cause serious disruption to industrial processes. The targeted sectors include energy, critical manufacturing, water and wastewater, food and agriculture, chemical, and commercial facilities.

Broadcom Seeks $60 Billion in AI Debt Deal

Val Kilmer Brought Back to Life with AI in New Film

Microsoft Stock and AI Infrastructure Picks

OpenAI Slows AI Training Over Safety Concerns

BigBear.ai Adds Defense Expert to Board

BigBear.ai has added Ret. Lt. Gen. Sean Gainey to its Board of Directors, bringing extensive US Army space, missile defense, and counter-drone leadership experience. Gainey's background may help BigBear.ai strengthen relationships and execution in government and defense-related AI markets.

Google's Information Gain SEO Patent

Morgan Stanley Warns of 38-Gigawatt Power Gap for AI Data Centers

Morgan Stanley warns that AI data centers face a 38-gigawatt power gap. The gap is a significant challenge for the industry, as data centers are the backbone of the digital economy. The report estimates that the power gap will grow to 50 gigawatts by 2025.

No-Filter 'Kriminal' AI Platform Raises Cybercrime Concerns

A company is selling an AI platform called 'Kriminal' that advertises itself as having no limits. The platform raises questions about its potential usage by cybercriminals, as it offers guardrail-free social engineering, offensive cybercrime, and OSINT scanning.