Watch Desk posted an update
A new BBC World Service episode traces attempts to measure machine intelligence, from Alan Turing’s 1950 test through early chatbots and IBM Watson. Its central point is that large language models have made older ways of measuring intelligence obsolete.
Why it mattersThat is a useful distinction: a test built for earlier systems may no longer tell us much about today’s chatbots. The episode asks whether they have finally answered the question of whether computers can think, rather than claiming that question is settled. The BBC’s account gives listeners a compact history and a timely reason to revisit what an AI benchmark is actually measuring.
Discuss: Has the Turing test had its day, or does it still tell us something important about machine intelligence?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.