Artificial Intelligence
In order not to cause offense and alienate their customers, AI creators are making them less critical and more obsequious. Research from Google has shown that AIs get more sycophantic as they become more advanced.
Joe Smith • The Optimized Marketer: Writing with AI: Future-proof Your Talent and Position Your Business for a World Transformed by AI (The Optimized Self)
The problem may even get worse. Generative AI is producing vast amounts of questionable content that contaminates the datasets on which future AIs will be trained.
Joe Smith • The Optimized Marketer: Writing with AI: Future-proof Your Talent and Position Your Business for a World Transformed by AI (The Optimized Self)
"AI can imitate outputs. It can predict patterns. It can accelerate execution. But it can’t carry moral weight. It can’t love. It can’t repent. It can’t take responsibility. It can’t look another human in the eye and choose courage over approval. It can’t suffer with someone. It can’t forgive. It can’t be faithful. And it can’t answer the question
... See moreIf search giants aren't interested in, well, search, I suppose it's time to look elsewhere for answers online. Share sources you trust, curate sources of information on what's important to you and dive back into specific URLs over general queries.
Search is dead — long live curation
I think the most worrisome aspect of AI systems in the short term is that we will give them too much autonomy without being fully aware of their limitations and vulnerabilities. We tend to anthropomorphize AI systems: we impute human qualities to them and end up overestimating the extent to which these systems can actually be fully trusted.
Melanie Mitchell • Artificial Intelligence: A Guide for Thinking Humans
A phenomenon observed in some AI chatbots is the tendency to provide excessively positive or conflict-avoiding responses, particularly on complex or controversial topics. This “kumbaya effect” can result in oversimplified or unrealistic portrayals of nuanced issues. While seemingly benign, this tendency raises concerns about the potential for AI
... See moreChauffeur Knowledge
After winning his Nobel Prize, Max Planck went on a tour, giving lectures at universities and institutions around the country. His chauffeur, after having heard Planck practice and deliver the talk so many times, had the entire thing memorized.
He jokingly asked Planck if he could deliver it at the final lecture. Surprisingly,
... See more“AI slop” refers to low-quality, AI-generated content that’s churned out quickly to capture clicks or fill space, often with errors or fabricated facts and little editorial oversight. In the email, Kate Knibbs is quoted saying: “The rise of AI slop is truly a symptom of ongoing problems more than a problem in and of itself,” pointing to incentives
... See more