AI is Deterministic Based Upon the Starting Data – AI Alignment Could Be Relatively Easy

An OpenAI employee has observed that Large Language Models starting with the same dataset converge to the same point. This would mean curating the data is the critical step in creating safe ASI (Artificial Super Intelligence). If we can front load the AI with the desired ethical and pro-humanity examples, then the resulting AI system …

Read more

DeepMind’s Shane Legg and Other Experts Talk AGI and AI Safety

Can we get AI large language models to rediscover human compatible (aligned) ethics ? Can we create AI world models that are pro-life, pro-humanity and pro-civilization. Can we create an AI that is an ethics and policy expert system? Paul Christiano (world’s leading AI safety researcher) talks about a post-AGI transition phase where current institutions …

Read more