September 15, 2026 | This week’s developments so far
Keeping up with artificial intelligence is starting to feel like watching a television series that releases a new season every afternoon. Today’s update covers four stories shaping the conversation this week, from rival AI companies discussing safety to a courtroom battle over what their systems can learn from.
For more headlines between these updates, check out my site The Drone Report, where AI news updates every half hour. Think of it as your AI news radar, minus the guy shouting that every new chatbot is going to change civilization by Thursday.
1. OpenAI and its rivals are reportedly talking safety
OpenAI has been discussing AI safety with Anthropic and Google DeepMind, according to Bloomberg reporting covered by Reuters on September 15. OpenAI policy chief Chris Lehane said those conversations have been underway for weeks. Reuters noted that it could not independently verify the report. The potential significance is cooperation among companies normally competing to outdo one another. My take: shared safety standards would be welcome, but the useful test will be what these companies actually commit to doing. An Avengers meeting is only helpful if someone leaves with an assignment. Read the reporting at Reuters.
2. Europe wants more control over its AI future
European Central Bank President Christine Lagarde warned on September 14 that Europe’s dependence on imported AI could leave essential services vulnerable if access changes or disappears. She called for more European computing capacity and models that run on local infrastructure. Her argument puts AI alongside other resources countries need reliable access to. My read: once hospitals, banks, and transportation systems depend on someone else’s technology, the subscription agreement becomes a geopolitical issue. “Have you tried restarting it?” is a weak national contingency plan. Read the story at Reuters.
3. OpenAI’s Indian copyright fight continues
The Delhi High Court declined on September 15 to immediately suspend an earlier ruling favorable to AI training in news agency ANI’s case against OpenAI. ANI argues that commercial training on copyrighted reporting should not qualify under exceptions for private use or research. The court requested OpenAI’s response and scheduled the next hearing for December 8. This is an ongoing dispute, not a final resolution of AI copyright questions. For writers and publishers, the stakes are familiar: who gets to use the work, who needs permission, and who gets paid? The robots have homework, and everyone is arguing over whose textbook they borrowed. Read the coverage at The Indian Express.
4. Bernie Sanders and Steve Bannon find common ground on AI
Here is a political crossover nobody ordered: Bernie Sanders and Steve Bannon were scheduled to speak at the same Washington gathering on September 15 to push for AI restrictions. Reuters reported the plans on September 14 amid concerns about jobs, household costs, and broader social harm. Sharing concerns does not mean sharing a complete policy agenda. Still, my takeaway is that AI oversight can scramble the usual political teams. We have apparently reached the buddy-movie portion of the technology debate. Whether this unlikely pairing produces legislation is the part worth watching. Read the story at Reuters.
Leave a Reply