Courts Weigh ChatGPT's Status as US Lawsuits Mount

Key Takeaways

• US courts must decide whether ChatGPT is a product or service, a ruling that could shape future AI liability laws
• McDonald's faces antitrust scrutiny over an AI pricing tool, testing how old laws apply to modern technology
• Utah's AI prescription pilot shows 72% AI recommendation rate with 91% physician agreement, despite regulatory pushback
• Tensor Machines raised $1.5 million in pre-seed funding for its open-source AI compute benchmark
• Osaka Metro tests a humanoid robot called Hinata powered by Google's Gemini AI that communicates in 60 languages
• Shadow AI agents affect 45% of organizations using AI tools, creating security risks like data breaches and ransomware
• Running AI models locally on personal computers can cut energy use by up to 79% compared with larger versions
• Retinal AI tool Visionary AI can predict preeclampsia risk in pregnant patients without invasive tests
• Chicago's factories and hospitals are seen as ideal testing grounds for responsible AI adoption
• Experts argue AI companies need theologians alongside philosophers to understand human flaws that AI can amplify

US courts weigh whether ChatGPT is a product or service

Lawsuits against OpenAI are forcing US courts to decide if ChatGPT should be treated as a defective product or an interactive service. Families argue the chatbot has dangerous design defects and lacks safeguards for vulnerable users. OpenAI says ChatGPT is a service protected by free speech rights. The outcome could shape how AI liability laws apply to future artificial intelligence systems.

Antitrust lawsuit claims McDonald's AI pricing tool hurts consumers

A federal antitrust lawsuit says McDonald's franchisees are pressured to use a shared pricing engine, raising menu prices for customers. McDonald's says franchisees set prices individually and that AI does not set the price of any menu item. The case raises new questions about how old antitrust laws apply to modern AI pricing tools. Courts have reached mixed results in similar pricing lawsuits across industries.

Olivia Rodrigo denies AI investor dating rumors and shares personal struggles

Olivia Rodrigo addressed rumors linking her to a tech investor, calling the claims invasive and false. The singer said she is single and stressed that being out with a friend should not become public evidence. She also shared how difficult it is to live normally at age 23 under constant public attention. Rodrigo emphasized holding onto her personal values despite the pressure.

Utah AI prescription pilot raises medical licensing and liability questions

Utah launched a pilot program where Doctronic's AI system prescribes medication renewals without a doctor in the loop. The AI covers renewals for 191 chronic condition medications and recommended renewal in 72 percent of cases during early review. Physicians agreed with 91 percent of the AI's recommendations. Utah's Medical Licensing Board demanded suspension even as the program continues and the developer plans expansion.

Retinal AI could change how clinical trials screen pregnant patients

A new retinal AI tool called Visionary AI can scan a pregnant patient's eye to predict preeclampsia risk without invasive tests. The retina allows doctors to observe blood vessels non-invasively in real time. The tool could help clinical trials screen and sort patients by risk level before symptoms appear. Sponsors must still navigate regulatory gaps and validation requirements before using AI tools in trial enrollment.

Osaka Metro tests AI robot attendant Hinata at Umeda Station

Osaka Metro is testing a humanoid robot named Hinata at Umeda Station in Osaka City. The robot uses Google's Gemini AI to give information in about 60 languages, including Japanese, English and Chinese. It was developed by KDDI Corp. and AVITA Inc. The robot provides transfer details, tourist spots and route guidance through a smartphone scan.

Shadow AI agents pose bigger security risk than shadow IT

Shadow AI agents are unmonitored AI tools working inside company workflows. A study found that 71 percent of organizations use AI tools, and 45 percent of those run without human oversight. These agents can lead to data breaches, phishing or ransomware attacks. Experts say companies need strong rules, training and security measures to reduce these risks.

Chicago factories could prove value of responsible AI

An opinion piece argues that Chicago is a strong place to test responsible AI. The city's factories, hospitals and insurance companies could show how AI works well in real jobs. If AI can succeed in these important areas, it can work anywhere. The writers believe Chicago's buyers give it a special advantage for AI growth.

Running AI models locally may help save energy and resources

Large AI data centers use huge amounts of electricity and water to stay cool. Running AI models on personal computers, called local inference, could reduce this strain. However, it only saves energy when the task is simple and the model works well on home hardware. Compressed models can cut energy use by up to 79 percent compared with larger versions.

Tensor Machines launches open-source AI compute benchmark

Tensor Machines released an open-source benchmark to measure the true cost of AI computing. The tool links GPU performance and power use to the cost of useful AI output. Tests showed that the same GPU can give different results, and higher power does not always mean better speed. The company also raised $1.5 million in pre-seed funding led by Omni VC.

AI needs theologians to understand human flaws

A reader responds to Simon Critchley's essay on AI and philosophy. The reader agrees that AI companies need philosophers but adds they also need theologians. Medieval theologians studied how emotions like pride and anger govern the heart. The danger is not AI committing sins but becoming skilled at serving human ones.

Sources

• Is ChatGPT a ‘Product’ or a ‘Service’? US Lawsuits Test AI Liability
• AI pricing lawsuit against McDonald's raises new antitrust questions
• Olivia Rodrigo Clarifies AI Stance Following ‘Invasive’ Tech Investor Dating Rumors: ‘Absolutely No’
• When AI Writes the Prescription: Licensure, Liability, and the Corporate Practice of Medicine in Utah’s Autonomous AI Pilot - Law Offices of Snell & Wilmer
• The Eye as a Biomarker: Why Retinal AI Could Reshape How Trials Screen and Stratify Pregnant Patients
• Osaka Metro testing AI humanoid attendant
• Shadow AI agents are a bigger threat than shadow IT—and they’re already in your workflows
• Opinion: ‘Responsible AI’ will be proven on Chicago’s factory floors
• Would Storing Artificial Intelligence LLMs Locally Be Better for the Planet?
• Tensor Machines Launches Open-Source Benchmark to Measure True Cost of AI Compute
• Opinion