Computers can control rockets without humans
SAFETY
-
LLMs Have Same Causal Power as Computers
By
–
LLMs have the same causal power as computers. Computers can fly rockets into your house. Books cannot.
-
Model Weights vs API Access: Information Extraction and Security
By
–
that’s a good point. with an API you can make lots of little updates, which providers definitely do what i meant was that people can learn a lot from the weights – info about the training process and tokenizer and data and architecture. stuff you can’t get from an API
-
Historical AI Doomsday Predictions Have Consistently Proven Wrong
By
–
These are the same kinds of shoddy, lazy, considerations that we've had many times before with resources, overpopulation, nuclear energy, particle colliders, biotechnology, GMOs, climate change, etc. And historically, EVERY SINGLE TIME the doomers were COMPLETELY wrong.
-
Elsevier combats scientific fraud with AI and human expertise
By
–
Safeguarding the vital work of #researchers requires constant rigor and vigilance. #Elsevier is investing heavily in human expertise and #tech to identify fraudulent content and prevent it from entering the scientific record. https://
bit.ly/4jbyKKz
#Research -
AI Training Infrastructure Challenges in Remote African Regions
By
–
The night our AI truly comes alive, the grid dies. 1/
We’re training a model in a remote area in South Africa.
Midnight. Stage 6.
Power fails. Data vanishes. Half the sensors go dark. In most labs, that’s a failed run.
In Africa, it’s Monday . -
Best Practices for AI Model Release Strategies
By
–
What do you think is the best approach for releasing models?
-
ChatGPT Privacy Risks: Five Critical Data Safety Warnings
By
–
Chat-GPT Danger: 5 Things You Should Never Tell The AI Bot #AI bot #ChatGPT has taken over our lives with a billion daily questions, but experts now warn it's a "privacy black hole" that could expose your most #personalinformation. This shocking exposé reveals exactly what you
-
Capability Denialists and Turing Machines: Cognitive Function Debate
By
–
Capability denialists tend to claim that a class of Turing Machines or universal learning algorithms cannot perform brain-like functions because of properties they don't like (not symbolic enough, too symbolic, too digital, not recurrent enough, not embodied, not social).
-
Denied AI Capabilities: Understanding Limitations and Constraints
By
–
The denied capabilities tend to be informally specified (understanding, intentionality, reasoning, agency, creativity, selfhood, sentience, sapience, abstraction, common sense) but can also apply to specific tasks (categorizing an object as the Charles river, winning at Go).