. @yLeCun continues to lie about my views on Threads, saying e.g. that I have argued that “Deep Learning [will not] remain the foundation of present and future AI systems” when I never said any such thing. What I have *actually* said over and over is deep learning is possibly
@garymarcus
-
Why Scientists Should Fear AI Development Risks
By
–
Everyone should be terrified. Not just scientists, but anyone who depends on science and well-trained scientists: tech companies, pharma companies, ordinary citizens, etc.
-
U.S. Healthcare and Science System Understanding Gap
By
–
If you are not liking this tweet you probably have almost no idea about how healthcare and science in the U.S. actually work. Or you just don’t care.
-
Poor Judgment Leader Nominated to Lead Healthcare and Medical Science
By
–
Man with catastrophically judgement nominates man with catastrophically bad judgement to “run wild with” healthcare and medical science.
-
AI Debate Recording with Mollick and Wales Tomorrow
By
–
Recording a debate on AI w @emollick and @jimmy_wales tomorrow – should be lots of fun!
-
Two Years Later: Deep Learning Progress and Limitations
By
–
https://
garymarcus.substack.com/p/two-years-la
ter-deep-learning-is
… -
Distinguishing Value from AGI Pathways in AI Development
By
–
Why are you directing that me? I didn’t say it wasn’t of value, I said it wasn’t in itself a path to AGI. It isn’t.
-
OpenAI Whisper’s hallucination problem: fabricated harmful content
By
–
“[OpenAI’s] Whisper has a major flaw: It is prone to making up chunks of text or even entire sentences… [S]ome of the invented text — known in the industry as hallucinations — can include racial commentary, violent rhetoric and even imagined medical treatments.
…
More -
LLM Safety Concerns in Medical Applications Highlighted
By
–
100% agree with @rajiinio
. Please do see also the specific alarm re LLM ls and medicine that she cites. -
AI Scaling Potential: From Current Capabilities to Exponential Growth
By
–
5 chicagos worth for now (if you believe the estimate) and maybe scaling takes it another 10, 100, 1000, 10,000x….