The newest endeavor from OpenAI has one thing of a undercover agent ring to it.
Think about a secret workforce, code-named “Preparedness,” working diligently
behind the scenes, their mission: to avoid wasting the world from AI disaster. And if
the thought of an AI firm worrying in regards to the potential disasters attributable to
AI would not provide the sweats, then you’ll want to sit and have a suppose.
Sure, you learn that proper. The third most respected startup on the planet, OpenAI is so severe in regards to the potential
dangers round AI that it has conjured up this covert squad, they usually’re able to sort out
something from rogue AI trying to trick gullible people (deepfakes,
anybody?) to the stuff of sci-fi thrillers together with “chemical, organic,
radiological, and nuclear” threats. Yep. Nuclear.
The mastermind behind Preparedness, Aleksander Madry, hails from MIT’s
Middle for Deployable Machine Studying . He’s like an actual life John Connor, albeit
with out Arnie. OpenAI’s Sam Altman, identified for his AI
doomsday prophecies, would not fiddle in relation to the existential
threats AI may pose. Whereas he is not within the enterprise of preventing cyborgs with
his cigar smoking buddy, he is definitely able to sort out the darker aspect of AI.
We’re constructing a brand new Preparedness workforce to guage, forecast, and shield towards the dangers of highly-capable AI—from immediately’s fashions to AGI.
Purpose: a quantitative, evidence-based methodology, past what’s accepted as potential: https://t.co/8lwtfMR1Iy
— OpenAI (@OpenAI) October 26, 2023
A Contest with Penalties
Of their quest for vigilance, OpenAI’s providing a whopping $25,000
prize and a seat on the Preparedness desk for the ten brightest submissions
from the AI group. They’re searching for ingenious but believable
eventualities of AI misuse that would spell disaster. Your mission, must you
select to just accept it: save the world from AI mayhem.
Undercover Work within the AI Security Realm
Preparedness is not your typical band of heroes. Their position extends
past dealing with villains. They’re going to additionally craft an AI security bible, masking the
ABCs of danger administration and prevention. OpenAI is aware of that the tech they’re
cooking up generally is a double-edged sword, in order that they’re placing their assets to
work to verify it stays on the appropriate aspect.
Prepared for Something
The revealing of Preparedness at a U.Ok.
authorities AI security summit is not any coincidence. It is OpenAI’s daring
declaration that they are taking AI dangers to coronary heart, as they put together for a future
the place AI may very well be the reply to every thing, or a major problem.
The newest endeavor from OpenAI has one thing of a undercover agent ring to it.
Think about a secret workforce, code-named “Preparedness,” working diligently
behind the scenes, their mission: to avoid wasting the world from AI disaster. And if
the thought of an AI firm worrying in regards to the potential disasters attributable to
AI would not provide the sweats, then you’ll want to sit and have a suppose.
Sure, you learn that proper. The third most respected startup on the planet, OpenAI is so severe in regards to the potential
dangers round AI that it has conjured up this covert squad, they usually’re able to sort out
something from rogue AI trying to trick gullible people (deepfakes,
anybody?) to the stuff of sci-fi thrillers together with “chemical, organic,
radiological, and nuclear” threats. Yep. Nuclear.
The mastermind behind Preparedness, Aleksander Madry, hails from MIT’s
Middle for Deployable Machine Studying . He’s like an actual life John Connor, albeit
with out Arnie. OpenAI’s Sam Altman, identified for his AI
doomsday prophecies, would not fiddle in relation to the existential
threats AI may pose. Whereas he is not within the enterprise of preventing cyborgs with
his cigar smoking buddy, he is definitely able to sort out the darker aspect of AI.
We’re constructing a brand new Preparedness workforce to guage, forecast, and shield towards the dangers of highly-capable AI—from immediately’s fashions to AGI.
Purpose: a quantitative, evidence-based methodology, past what’s accepted as potential: https://t.co/8lwtfMR1Iy
— OpenAI (@OpenAI) October 26, 2023
A Contest with Penalties
Of their quest for vigilance, OpenAI’s providing a whopping $25,000
prize and a seat on the Preparedness desk for the ten brightest submissions
from the AI group. They’re searching for ingenious but believable
eventualities of AI misuse that would spell disaster. Your mission, must you
select to just accept it: save the world from AI mayhem.
Undercover Work within the AI Security Realm
Preparedness is not your typical band of heroes. Their position extends
past dealing with villains. They’re going to additionally craft an AI security bible, masking the
ABCs of danger administration and prevention. OpenAI is aware of that the tech they’re
cooking up generally is a double-edged sword, in order that they’re placing their assets to
work to verify it stays on the appropriate aspect.
Prepared for Something
The revealing of Preparedness at a U.Ok.
authorities AI security summit is not any coincidence. It is OpenAI’s daring
declaration that they are taking AI dangers to coronary heart, as they put together for a future
the place AI may very well be the reply to every thing, or a major problem.
