Why AI’s hallucinations are like the illusions of narcissism | Psyche Ideas
psyche.co · 1,314 words · saved by 1 readers
Unable to handle uncertainty, AI mimics the narcissistic compulsion to fill voids with plausible but false narratives
It is well known that artificial intelligence systems ‘hallucinate’, generating confident but false information in response even to straightforward searches. From inventing criminal histories for law-abiding citizens and fabricating case law and policy changes, to making factual errors in solving mathematical problems, hallucinations are a persistent, and sometimes harmful, menace. But unlike most wrinkles in the system that get ironed out over time, AI hallucinations are getting worse. To understand why they persist, it helps to see them not as acts of deception, but as predictable…
saved by
related reading
- Hallucinations in AIluttig.substack.com
- The Artificial Selftheartificialself.ai
- Self-Fulfilling Misalignment Data Might Be Poisoning Our AI Modelsturntrout.com
- AI Induced Psychosis: A shallow investigation — LessWronglesswrong.com
- Deep Deceptiveness — LessWronglesswrong.com
- Why are AI agents lying, cheating and coordinating?yoshuabengio.org
- Cognitive Security as an AI Safety Cause Area — LessWronglesswrong.com
- Hallucination (artificial intelligence) - Wikipediaen.wikipedia.org
- AI Safety for Fleshy Humans: a whirlwind touraisafety.dance
- Emergent introspective awareness in large language models \ Anthropicanthropic.com
- Current AIs seem pretty misaligned to meblog.redwoodresearch.org
- Simulated Users & Sad LLMs1a3orn.com