AGI stands for artificial general intelligence: a machine that could do <b>any</b> intellectual task a person can, rather than being good at one narrow thing. It is the thing people mean when they ask whether AI will replace everyone. The honest answer has three parts.
Today's systems are extraordinary at language and specific tasks, and still fail at things a careful ten-year-old would not: reliable arithmetic, knowing what they do not know, and noticing when a question makes no sense.
They also have no way of being responsible for anything, which is not a small missing feature.
Serious researchers give estimates ranging from a few years to never, which tells you the honest answer is that this is not predictable.
Be suspicious of anyone who is certain in either direction — both "it is imminent" and "it is impossible" are being sold to you by someone.
It does hours of real work: reading, drafting, sorting, explaining, translating.
It cannot hold responsibility for any of it. Responsibility is not a capability that arrives in the next version — it belongs to a person by definition.
The useful mental model is not "a replacement". It is a very fast, very well-read junior colleague who has never once said "I'm not sure".
Firmly in the middle band, and moving. The gap that matters is accountability, not capability.
The last activity in our free course asks you where we are today, shows you where everyone else put it, then gives the honest answer.
▶ Play Lemonade Empire free