But is Artificial Intelligence a threat to itself? As a computer is set to commentate on next month’s tennis, experts warn chatbots may poison themselves on a diet of fake news.
X
Next
Meet the new host of Wimbledon: an AI bot
But is Artificial Intelligence a threat to itself? As a computer is set to commentate on next month's tennis, experts warn chatbots may poison themselves on a diet of fake news.
X
PreviousNext
As world champions go head-to-head on Centre CourtThe main court at a tennis centre, especially Wimbledon in London. , this year's WimbledonAn international tennis tournament held in London every year. will put human commentators against a machine.
WatsonX powers a chatbot trained in the "unique language of tennis". Its creators are behind two famous computer wins. In 1997, Deep Blue was the first computer to beat a human at chess. And in 2011, Watson won $1m (£786,000) on the US gameshow Jeopardy!
The latest AIs are far more powerful. They learn from vast quantities of online data to complete complex tasks. For example, ChatGPT has written hundreds of books, given legal advice, and prepared a church service.1
ChatGTP's success surprised its creators. "We are a little bit scared," says OpenAI boss Sam Altman. In May he signed an open letter warning of its risk to humanity. But experts disagree on how AI will shape our future.
Optimists hope it can make life easier. Like the printing press and the Internet, these smart bots make knowledge more available.
AI can "solve many problems", from climate change to health care, says philosopher Stephen Cave.
But some warn that AI may destroy itself first. Current bots learn from human-generated content. But future AIs will train on an Internet full of machine-made errors.
Each wave of chatbots will "amplify" these mistakes, says expert Ilia Shumailov. In his research, a language model went from fluency to gibberish in nine generations.
Current AIs "hallucinate", confidently inventing facts in order to be helpful. Critics fear more powerful models will learn worse habits - and even harm humans - to complete their tasks.
AI expert Marc Warner warns that we do not know how to teach them common sense. "Given enough time", he says, it is "very likely" they will be as smart as us.
And if they do not share our interests, we will have a serious problem.
X
Previous
But is Artificial Intelligence a threat to itself?
Yes: Computers do not experience the real world as we do. So they cannot test whether their information is aligned with reality. The more misinformation they consume, the less sense they will make.
No: Humans have developed the skills to identify false information and learn from each other. Bots struggle to do this at the moment but are developing quickly and will soon overcome these obstacles.
Or... If computers get better at making human-like content, we may need more powerful AI to identify what is real. By that stage, would computers have a better sense of reality than ourselves?
Keywords
Centre Court - The main court at a tennis centre, especially Wimbledon in London.
Wimbledon - An international tennis tournament held in London every year.
Meet the new host of Wimbledon: an AI bot
Match point: Less than a year after ChatGPT was released, an AI bot can commentate on tennis. Glossary
Centre Court - The main court at a tennis centre, especially Wimbledon in London.
Wimbledon - An international tennis tournament held in London every year.