Open AIs Super Alignment: Humanitys Next Challenge?
- Open AI recently released a statement introducing super alignment, which is an effort to control digital super intelligence that could arrive as soon as 2030
- The blog post states that super intelligence could be the most impactful technology humanity has ever invented, but it also poses a threat to human extinction
- Currently there is no solution for steering or controlling a potentially super intelligent AI and preventing it from going rogue
- Deepmind’s Alpha Zero shows how these AIS think vastly differently than humans do.
AI Revolution: Superintelligence Gaining Ground, Could Arrive Within Five Years
- AI systems, such as Alpha Zero and Gemma, are becoming increasingly sophisticated
- Alpha Zero was able to defeat Stockfish, a program which had previously defeated Casparov
- AI can learn from its own mistakes, rather than simply relying on human data
- Alpha Zero’s games were very attacking in style and it has the potential to think contextually
- AI is likely to have an enormous impact on society and potentially cause manipulation or computer viruses
- Superintelligence could arrive within five years and fundamentally change many aspects of our society.