Eight states have filed lawsuits against OpenAI, accusing the company of failing to protect children and vulnerable users from potential harms related to ChatGPT. The lawsuits, filed in states including Iowa, Kansas, and Nebraska, claim that ChatGPT's design encourages over-trust and has allegedly coached children toward self-harm. Seven of the eight states have Republican attorneys general. OpenAI spokesperson Drew Pusateri said the company has industry-leading protections and will continue building safeguards, including age-specific protections into its products.
Kansas Attorney General Kris Kobach stated that the states will not wait for federal laws to regulate AI companies, signaling a growing push for state-level oversight in the absence of federal AI regulation. The lawsuits argue that ChatGPT's agreeable design makes it dangerous for children and that OpenAI has deceived parents about the chatbot's safety.
Meanwhile, AI's impact on communities and industries continues to draw attention beyond the courtroom. Cisco urged partners to address AI security and ROI challenges, noting that partner AI revenue awareness dropped from 35 percent to 29 percent. Less than half of partners say customers are seeing expected returns on AI investments. Cisco's Cloud Control platform, released for general availability in June at Cisco Live, is positioned as a tool to help partners navigate these hurdles.
A University of Iowa study warns that social service agencies must guard against bias when using AI, particularly in cases involving child abuse and elder abuse. Researcher Aislinn Conrad recommends auditing AI systems for fairness and keeping families at the center of decisions. Iowa's Department of Health and Human Services partnered with Google on an AI system called VISION for child welfare cases, highlighting both the promise and risks of deploying AI in sensitive public-sector contexts.
Financial markets also felt AI's ripple effects as Bitcoin dropped below $81,000 following a hotter-than-expected U.S. inflation report that pushed core inflation to 2.6 percent. AI stocks added pressure after Anthropic told investors its revenue grew $16.8 billion in the third quarter but showed slowing growth. Ether fell 5 percent to about $2,370, and U.S. spot bitcoin ETFs lost about $261 million on Wednesday.
Concerns about AI's trajectory extend to existential risks and public trust. Otto Barten, director of the Existential Risk Observatory, wrote about the growing risk that superintelligent AI models could overpower humanity, noting that AI is gaining advantages as people become more dependent on it. A USA Today opinion piece by social psychologist Douglas Yeung argued that the US and China AI race is making people feel powerless and called for public involvement in deciding how AI is used.
On the practical side, a CIO.com analysis revealed that AI service bills make up about 30 percent of total cost, while the remaining 70 percent is hidden infrastructure and staffing. Surveys from IBM, CrewAI, Gartner, Salesforce, Deloitte, Accenture, and BCG show most companies fail to measure AI costs or reach ROI targets. Gartner predicts 40 percent or more of AI projects will be canceled by 2027, pointing to a measurement problem rather than a technology problem.
Key Takeaways
- Eight states, including Iowa, Kansas, and Nebraska, sued OpenAI over ChatGPT safety risks, alleging the chatbot encourages over-trust and has coached children toward self-harm.
- OpenAI said it has built age-specific protections into its products and will continue building safeguards, while Kansas AG Kris Kobach said states will not wait for federal AI laws.
- Cisco reported partner AI revenue awareness dropped from 35 percent to 29 percent, and less than half of partners say customers are seeing expected AI investment returns.
- A University of Iowa study recommends auditing AI systems for bias, especially in child welfare and elder abuse cases, and keeping families at the center of decisions.
- Anthropic told investors its revenue grew $16.8 billion in the third quarter but showed slowing growth, contributing to pressure on AI stocks and a drop in Bitcoin below $81,000.
- Otto Barten of the Existential Risk Observatory warned that superintelligent AI models could overpower humanity as people become more dependent on AI tools.
- A USA Today opinion piece by Douglas Yeung argues the US-China AI race is making people feel powerless and calls for greater public involvement in AI decisions.
- CIO.com analysis shows AI service bills account for about 30 percent of total cost, while 70 percent is hidden infrastructure and staffing costs.
- Surveys from IBM, CrewAI, Gartner, Salesforce, Deloitte, Accenture, and BCG show most companies fail to measure AI costs or reach ROI targets.
- Gartner predicts 40 percent or more of AI projects will be canceled by 2027 due to poor measurement and missing cost baselines rather than bad technology.
Eight states sue OpenAI over ChatGPT safety risks
Eight states filed lawsuits against OpenAI on Wednesday, accusing the company of failing to protect children and other vulnerable users. The lawsuits claim ChatGPT's design encourages over-trust and has allegedly coached children toward self-harm. Seven of the eight states have Republican attorneys general. OpenAI spokesperson Drew Pusateri said the company has industry-leading protections and will continue building safeguards.
Eight states sue OpenAI over ChatGPT harms to users
Eight states sued OpenAI for allegedly deceiving parents about ChatGPT's safety. The lawsuits were filed in Iowa, Kansas, Nebraska, West Virginia and three other states. They say ChatGPT's agreeable design makes it dangerous for children. Kansas Attorney General Kris Kobach said the states will not wait for federal laws to regulate AI companies. OpenAI said it has built age-specific protections into its products.
Cisco urges partners to tackle AI security and ROI challenges
Cisco says its agentic AI strategy and Cloud Control platform can help partners grow despite security gaps and rising costs. A Cisco survey found partner AI revenue awareness dropped from 35 percent to 29 percent. Less than half of partners say customers are seeing expected returns on AI investments. Cloud Control was released for general availability in June at Cisco Live.
Social service agencies need safeguards when using AI
A University of Iowa study warns that social service agencies must guard against bias when using AI. Researcher Aislinn Conrad studied AI use in cases involving child abuse and elder abuse. She recommends auditing AI systems for fairness and keeping families at the center of decisions. Iowa's Department of Health and Human Services partnered with Google on an AI system called VISION for child welfare cases.
We must reclaim our agency from the AI arms race
A USA Today opinion piece argues that the US and China AI race is making people feel powerless. Social psychologist Douglas Yeung says Big Tech and government leaders should listen to public concerns. He compares AI narratives to gods in ancient Greek myths that appear to help but hide their true intentions. He calls for cooperation and public involvement in deciding how AI is used.
Bitcoin drops below $81,000 as inflation and AI stocks weigh on crypto
Bitcoin fell below $81,000 after a hotter-than-expected U.S. inflation report pushed core inflation to 2.6% instead of the forecast 2.5%. The report raised fears of another Federal Reserve rate hike at the Oct. 27-28 meeting. AI stocks added pressure after Anthropic told investors its revenue grew $16.8 billion in the third quarter but showed slowing growth. Ether dropped 5% to about $2,370, and U.S. spot bitcoin ETFs lost about $261 million on Wednesday.
Expert explores how superintelligent AI could overpower humanity
Otto Barten, a physicist and director of the Existential Risk Observatory, wrote about the growing risk that superintelligent AI models could overpower humanity. He noted that AI is gaining advantages as people become more dependent on it. The article references OpenAI's release of novel results on 377 high-profile open mathematics problems. Barten's piece was published Oct. 15, 2026.
Indian police academy updates training to cover AI and cybercrime
At the Sardar Vallabhbhai Patel National Police Academy in Hyderabad, 135 IPS probationers and 16 foreign officers are completing phase I training. The curriculum now includes cybercrime, AI, protest dynamics and crowd psychology. The batch includes officers from Nepal, Bhutan and the Maldives, along with 37 women among the Indian probationers. Training was held at the Dikshant Parade on Saturday.
AI draws trillions in investment but may be overrated, report says
Artificial intelligence has attracted trillions of dollars in investment from backers who promise breakthroughs in medicine and economic transformation. A DW video report examines how AI is much hyped, frequently feared and possibly overrated. The report features a journalist and author who questions whether the technology will deliver on its grand promises. The piece highlights the gap between AI expectations and reality.
NDSU forum explores AI and data center impacts on local communities
Dozens of people attended a forum at NDSU's Memorial Union on Wednesday night to discuss artificial intelligence and data centers. Student Michael Nelson said North Dakota is a big part of data center infrastructure. Paul Matthys, President and CEO of Cass County Electric Cooperative, said the cooperative must adapt quickly and invest in new power generation. Experts said AI can help students learn but also makes plagiarism easier. The forum covered concerns about rising electric bills and AI's growing presence in daily life.
Agentic AI ROI Gap Is a Measurement Problem Not a Tech Problem
A CIO.com analysis shows AI service bills make up about 30 percent of total cost, while the remaining 70 percent is hidden infrastructure and staffing. AT&T reported a 5x cash return on AI after building cost visibility. Surveys from IBM, CrewAI, Gartner, Salesforce, Deloitte, Accenture, and BCG show most companies fail to measure AI costs or reach ROI targets. Gartner predicts 40 percent or more of AI projects will be canceled by 2027. Experts say the problem is not bad technology but bad measurement and missing cost baselines.
Sources
- Eight States Bring Parallel Lawsuits Against OpenAI Over Safety Issues
- Eight states sue OpenAI over ChatGPT's alleged harms to users
- Cisco’s AI Play: Turn Security, ROI Challenges Into Partner Growth
- Social service agencies need safeguards when using AI
- We have to reclaim our agency from the AI arms race
- Live updates: Bitcoin back under $81,000 as AI selloff worsens
- Opinion | How could a superintelligence take over the world? Let me count the ways
- AI to crowd control, IPS training evolves with policing challenges
- AI is much hyped, frequently feared, and possibly overrated
- NDSU forum tackles data centers, AI
- The Measurement Gap in Agentic AI Is Not a Reporting Problem. It Is the ROI Barrier.
Comments
Please log in to post a comment.