What do YOU think?
We have questions. You have answers.
The future is coming too fast! (By which we mean 2026.)
But fear not. We’re hatching great reads for your new year: Can AI help fix government?; What exactly is “AI manipulation”?; and Might sci-fi hold clues about the world we’re hurtling towards?
All we lack is you. More exactly, your intelligence on artificial intelligence. So……
Please complete this quick (5 minutes?) questionnaire.
Given plunging survey response rates, we’ve limited ourselves to just 2 questions. Write 2 words, or 200.
What’s an AI topic you’d like better explained?
What’s a topic that people aren’t discussing enough?
We’ll read every answer with great interest. Wishing you an excellent 2026!
—Conor Griffin & Tom Rachman, AI Policy Perspectives
I'd like a better explanation of the "hallucination" issue with LLM-based chatbots. I've read that researchers are having a tough time figuring out how to solve it, either because they don't fully understand why it happens or because it is inherent in the way they work. Where do we stand on this? Will trust in AI be limited to our level of trust in the Internet at large?