Sascha Brodsky

AI communications @IBM. Journalist with bylines in The New York Times, LA Times, Time, the Guardian, The Atlantic, Reuters and more
[email protected]

[Sascha Brodsky @BrodskySascha]

Bad AI advice may turn doubt into false confidence.

A new paper suggests that AI can make people less willing to admit when they don’t know an answer. In experiments, incorrect AI advice made participants less accurate but more confident. https://t.co/LTra3FIExr

[Sascha Brodsky @BrodskySascha]

@IBM and @Google Cloud have launched a new strategic partnership aimed at helping enterprises move AI from pilots into production. https://t.co/BrK6E7WOoV

[Sascha Brodsky @BrodskySascha]

When one AI model isn't enough https://t.co/ByuV2Xnq5X

[Sascha Brodsky @BrodskySascha]

The Pentagon’s AI Problem https://t.co/4TnBoyg1Js

[Sascha Brodsky @BrodskySascha]

AI was supposed to buy us time. Instead, it is taking it.
A new study from Harvard Business Review found that enterprise AI made employees work faster, take on broader responsibilities, and log longer hours without being asked.

[Sascha Brodsky @BrodskySascha]

In my interview with James Muldoon on Love Machines:
We worry about AI taking jobs.
We ignore it taking up emotional space.
Link ↓https://t.co/ASW4e5WZeI

[Sascha Brodsky @BrodskySascha]

We’re told generative AI will unlock creativity. New research suggests it may also make us think more alike.

https://t.co/6Fl8HHrg0C

[Sascha Brodsky @BrodskySascha]

AI isn’t eliminating most jobs. It’s splitting them in two.
Those who learn to work with AI will surge ahead. Those who don’t will be left competing in a shrinking middle.
My deep dive for IBM on what AI really means for the future of work: https://t.co/SMx6nuUTrX

[Sascha Brodsky @BrodskySascha]

Just published: The AI world is quietly ditching leaderboards. 🧠

"You end up with a model that seems smart, but has memorized the test," @IBM Gabe Goodhart told me.

Story 🔗https://t.co/Xd3DLFzu6f
#AI #MachineLearning #IBM

[Sascha Brodsky @BrodskySascha]

🚨 Just launched: Quantum AI Brief — my free weekly newsletter decoding the intersection of quantum computing and artificial intelligence.
https://t.co/V9aDUNrV7V

[Sascha Brodsky @BrodskySascha]

@Thom_Wolf I agree with all this, but can't AI just overwhelm science problems with volume rather than creativity?

[Sascha Brodsky @BrodskySascha]

Just published my latest piece for @IBM Think: "DeepSeek shows there's room for more AI players"

@ArvidKrishna: "There is no law of physics that dictates AI models must remain big and expensive"

https://t.co/EA2puXgCAB
#AI #Innovation #Technology #IBMResearch

[Steven Adler @sjgadler]

Some personal news: After four years working on safety across @openai, I left in mid-November. It was a wild ride with lots of chapters - dangerous capability evals, agent safety/control, AGI and online identity, etc. - and I'll miss many parts of it.

[Sascha Brodsky @BrodskySascha]

Check out the latest article in my newsletter: AI Game-Changer: DeepSeek Challenges Industry Assumptions with $6M Breakthrough https://t.co/5CjNFwK2EN via @LinkedIn

[Matt Rosoff @MattRosoff]

Some personal news: TechCrunch and I have decided to part ways. It's been a learning experience and I met a lot of great folks there. If anybody is looking for an experienced newsroom leader with a history of training up young reporters to break news, give me a shout.

[Sascha Brodsky @BrodskySascha]

I am seeking an AI expert with expertise in world models to interview for a forthcoming article; the deadline is tomorrow. Get in touch at [email protected]

[Sascha Brodsky @BrodskySascha]

Check out the latest article in my newsletter: AI News, October 13, 2024 https://t.co/LUZg6ab5h6 via @LinkedIn

[Jason Wei @_jasonwei]

o1-mini is the most surprising research result i've seen in the past year

obviously i cannot spill the secret, but a small model getting >60% on AIME math competition is so good that it's hard to believe

congrats @ren_hongyu @shengjia_zhao for the great work!

[clem 🤗 @ClementDelangue]

Once again, an AI system is not "thinking", it's "processing", "running predictions",... just like Google or computers do.

Giving the false impression that technology systems are human is just cheap snake oil and marketing to fool you into thinking it's more clever than it is.

[Sascha Brodsky @BrodskySascha]

Are we seeing the beginning of AGI?
OpenAi's https://t.co/EFkgagqebz
https://t.co/EFkgagqebz

[Sam Altman @sama]

here is o1, a series of our most capable and aligned models yet:

https://t.co/yzZGNN8HvD

o1 is still flawed, still limited, and it still seems more impressive on first use than it does after you spend more time with it.