Is it possible to create an impartial AI system? The “godfather of AI” working at Meta says “absolutely not”
Loading... Loading...
Meta Platforms Inc. META chief AI scientist And "Godfather of AI" Yann LeCun has rejected THE possibility of create A impartial AI system.
What Arrived: Last week, In A episode of the Lex Friedman Podcast, LeCun discussed THE concept of "I woke up AI”, A term used has describe AI systems that are too much sensitive has social problems.
LeCun address THE critical of that of Google Gemini, accused of be "great wake up " And manufacturing historically inaccurate pictures such as A black George Washington, And refusing has generate pictures or descriptions of Tiananmen Square Square or the Reservoir Man, which are heavily censored by THE Chinese government.
LeCun argued that he East impossible has created a AI system that is not biased, as bias East subjective And varied Since person has person. He suggested that THE solution East has to have A miscellaneous range of AI systems, similar has THE diversity of opinions In A liberal democracy.
See Also: Elon Musk Slams AOC After new York Representative Makes 'Election Interference' Allegations About X: 'I Do THE Algorithm Open Source And Neutral'
"East he possible has produce A AI system that East not biased? And THE answer East, absolutely not. And It is not because of technological challenges, although they are technological challenges has that, It is because bias East In THE eye of THE spectator", LeCun said.
He went on has add, » Different people can to have different ideas about What constitutes bias For A plot of things, there are facts that are indisputable, but there are A plot of opinions Or things that can be Express In different manners. And SO You can't to have A impartial system, It is just A impossibility. »
LeCun Underlines THE importance of Open source AI systems, declaring that THE future of human interaction with THE digital the world is going be publicized by AI. He warned against A future Or A little companies on THE West Side of THE WE. control all AI systems, stressful THE need For diversity In AI development.
"I talked has THE French government enough A little, And THE French government will not accept that THE digital diet of all their citizens be control by three companies on THE West side of THE WE. It is just not acceptable, It is A hazard has democracy regardless of how well intentioned those companies are, And It is Also A hazard has local culture, has values, has language," he said.
Fridman summary her notice by saying, "SO meaning A business as Meta Or Google Or SO on should take only minimal fine tuning not After building THE foundation pre-trained model as little not as possible," has which LeCun replied by saying, "Basically."
For what He Imported: THE issue of bias In AI has has been A hot subject In THE technology industry. that of Google Gemini AI, For example, has confronted critical For generate historically inaccurate pictures, leading has A internal investigation and efforts...
Meta Platforms Inc. META chief AI scientist And "Godfather of AI" Yann LeCun has rejected THE possibility of create A impartial AI system.
What Arrived: Last week, In A episode of the Lex Friedman Podcast, LeCun discussed THE concept of "I woke up AI”, A term used has describe AI systems that are too much sensitive has social problems.
LeCun address THE critical of that of Google Gemini, accused of be "great wake up " And manufacturing historically inaccurate pictures such as A black George Washington, And refusing has generate pictures or descriptions of Tiananmen Square Square or the Reservoir Man, which are heavily censored by THE Chinese government.
LeCun argued that he East impossible has created a AI system that is not biased, as bias East subjective And varied Since person has person. He suggested that THE solution East has to have A miscellaneous range of AI systems, similar has THE diversity of opinions In A liberal democracy.
See Also: Elon Musk Slams AOC After new York Representative Makes 'Election Interference' Allegations About X: 'I Do THE Algorithm Open Source And Neutral'
"East he possible has produce A AI system that East not biased? And THE answer East, absolutely not. And It is not because of technological challenges, although they are technological challenges has that, It is because bias East In THE eye of THE spectator", LeCun said.
He went on has add, » Different people can to have different ideas about What constitutes bias For A plot of things, there are facts that are indisputable, but there are A plot of opinions Or things that can be Express In different manners. And SO You can't to have A impartial system, It is just A impossibility. »
LeCun Underlines THE importance of Open source AI systems, declaring that THE future of human interaction with THE digital the world is going be publicized by AI. He warned against A future Or A little companies on THE West Side of THE WE. control all AI systems, stressful THE need For diversity In AI development.
"I talked has THE French government enough A little, And THE French government will not accept that THE digital diet of all their citizens be control by three companies on THE West side of THE WE. It is just not acceptable, It is A hazard has democracy regardless of how well intentioned those companies are, And It is Also A hazard has local culture, has values, has language," he said.
Fridman summary her notice by saying, "SO meaning A business as Meta Or Google Or SO on should take only minimal fine tuning not After building THE foundation pre-trained model as little not as possible," has which LeCun replied by saying, "Basically."
For what He Imported: THE issue of bias In AI has has been A hot subject In THE technology industry. that of Google Gemini AI, For example, has confronted critical For generate historically inaccurate pictures, leading has A internal investigation and efforts...
What's Your Reaction?
