Sam Altman's 2025 AGI Outlook
Key Points
- Sam Altman’s New Year’s Reflections predict the arrival of artificial general intelligence (AGI) in 2025, specifically in the form of AI agents that act as colleagues in tools like Slack.
- These “AI coworkers” are expected to perform enough work to be billed at roughly ten percent of an equivalent employee’s salary, but they will still require human oversight and cannot replace entire organizations.
- Altman foresees a short‑term shift where companies hire these agents for specific tasks, causing minimal disruption to the broader labor market.
- He also signals a strategic pivot for OpenAI: half the business will continue offering AGI‑powered products (e.g., ChatGPT), while the other half will refocus on core research aimed at achieving superintelligence.
- The ultimate goal of that superintelligence, according to Altman, is to accelerate scientific breakthroughs—potentially solving hard problems like aging, cancer, and other unsolved diseases.
Full Transcript
# Sam Altman's 2025 AGI Outlook **Source:** [https://www.youtube.com/watch?v=7fJSM4G9Efg](https://www.youtube.com/watch?v=7fJSM4G9Efg) **Duration:** 00:04:29 ## Summary - Sam Altman’s New Year’s Reflections predict the arrival of artificial general intelligence (AGI) in 2025, specifically in the form of AI agents that act as colleagues in tools like Slack. - These “AI coworkers” are expected to perform enough work to be billed at roughly ten percent of an equivalent employee’s salary, but they will still require human oversight and cannot replace entire organizations. - Altman foresees a short‑term shift where companies hire these agents for specific tasks, causing minimal disruption to the broader labor market. - He also signals a strategic pivot for OpenAI: half the business will continue offering AGI‑powered products (e.g., ChatGPT), while the other half will refocus on core research aimed at achieving superintelligence. - The ultimate goal of that superintelligence, according to Altman, is to accelerate scientific breakthroughs—potentially solving hard problems like aging, cancer, and other unsolved diseases. ## Sections - [00:00:00](https://www.youtube.com/watch?v=7fJSM4G9Efg&t=0s) **Sam Altman's 2025 AGI Forecast** - Sam Altman predicts artificial general intelligence will appear in 2025 as AI agents that function as workplace colleagues—costing roughly 10% of an employee’s salary, causing limited short‑term disruption and prompting a shift in business focus. ## Full Transcript
you know when normal people write their
New Year's resolutions and their
Reflections nobody pays attention but
when Sam Altman does founder of open AI
everybody does so he wrote his new
year's Reflections I read through them
for you long story short Sam sees
artificial general intelligence this
year in 2025 and specifically he named
The Format he thinks it will take which
I think is important he hasn't said that
publicly before he says it's going to
look like AI AG agents that are
colleagues in slack AI agents that are
colleagues at work that you can partner
with to get jobs
done that's it that's that's what he
defined as important enough and he
thinks that they will do good enough
work that companies will pay for them
now if I were to guess he did not say
this but if I were to guess I would say
typically that's going to get priced in
about 10% of the equivalent employee
salary and before everyone says well
that's the end of jobs as we know it I
don't think so because these agents by
definition can't do the work of an
organization yet and so they do need
humans around them above them below them
Etc and so I think what we're going to
see is just some new AI colleagues and
that's what it's going to look like and
it will actually be fairly minimal
disruption in the short term at the same
time Sam mman shared something else
that's interesting he thinks that the
focus of the entire business is going to
shift from on the one hand running these
sort of generally intelligent agents as
a product maybe running Chad GPT as a
product with one half of the business
and then the other half is going to go
back to what open AI has traditionally
focused on which is research and
specifically research towards super
intelligence and this is something that
over the holiday break various open AI
Associated Affiliated working at kind of
people were saying we see something we
think that super intelligence is near
this is sort of Sam's way of saying he
sees a path to Super intelligence and
typically super intelligence has been
defined as an AI that can do the work of
an entire organization not just one
person but interestingly enough when Sam
talked about it in his Reflections that
wasn't the definition he focused on he
might not disagree with it but what he
focused on was the power of super
intelligence to accelerate scientific
research and I think he's really excited
because he sees some of our biggest
hardest problems as a species as
something that super intelligence could
help us solve so the classic one is
aging what if super intelligence could
find a solution for aging and then we
wouldn't have to age and
die right like that's as old as like
Humanity's quest to to avoid death right
like that's not new uh but Sam is
thinking that super intelligence might
be able to help with that there's a
bunch of other ones right like uh long
unsolved diseases quote unquote the cure
for cancer sir um and so I think what
Sam is interested in and where he hinted
that open AI is going is that as they
get closer to Super intelligence they're
going to Pivot more into a scientific
research lab and partner more with
scientific institutions I think you saw
a hint of that in the holiday
announcements when they tipped their cap
a few times on a few of those 12 days of
open AI to Scientific institutions into
how they help researchers I think they
want to do that more um I think they
want to build closer partner closer
partnership links I think they want to
do some scientific research themselves
and I think that's something that Sam
sees as a long-term Legacy for the
company if he can solve some of these
big problems with super intelligence he
will feel like the work that he did had
some meaning to it and that's something
that in in our small way that's that's
what we all want right uh and it's very
appropriate for a New Year's reflection
so there you go uh Sam thinks that
general intelligence is coming this year
and Sam also thinks that uh super
intelligence is the next big project
that open AI has he did not put a
timeline on that one by the way so we'll
see that by the way is my first
prediction from 2025 that I got right I
was very pleased with myself so I I I
was guessing that we were going to get
uh agents uh like in slack collaborating
with our with us as colleagues um so I
need to take one all the others I could
still get wrong but I'll take one uh I I
put that up on my substack it was fun to
see that one come right in the first few
days of the year all right cheers