Yoshua Bengio: Democracy is not safe in an AI world | Talk to Al Jazeera
Yoshua Bengio's work in the 1990s and 2000s has contributed to the foundations of chatbots like OpenAI's ChatGPT and Google's Bard.
Today, Bengio is one of the biggest voices warning the world about the necessity of having control and regulations on artificial intelligence (AI) technology.
He testified before a United States congressional hearing, warning that the frantic pace of AI development, in the wrong hands, could be used to create biological weapons.
So, is AI an existential threat to humanity? Or is it a tool that will transform our lives for the better?
Computer Scientist Yoshua Bengio talks to Al Jazeera.
Subscribe to our channel bit.ly/AJSubscribe
Follow us on Twitter / ajenglish
Find us on Facebook / aljazeera
Check our website: www.aljazeera.com/
Check out our Instagram page: / aljazeeraenglish
Download AJE Mobile App: aje.io/AJEMobile
@AljazeeraEnglish
#Aljazeeraenglish
#News
We, as the human species, are competitive by nature. I believe the world is becoming smaller every day, so we need to find a target beyond this small planet. This will allow us all to say, "This is our home," and we can start cooperating with each other to reach that external goal from this planet - our home. However, until we can realize this, it doesn't matter what we do. AI, in this context, is just a thing. We can compare it to industrialization 100 years ago, but my argument is that it is just a thing that makes the world smaller again. I believe that development cannot be stopped, so I think the very smart people, like the interviewee, should start looking beyond this small planet with his colleagues - I am pretty sure they are very smart too. They should find a goal, a target, a purpose outside of this small planet, so we can start cooperating and not developing weapons to kill ourselves at "home." Maybe it would be great if we could find some proof that we are not alone in this universe, but that idea is far-fetched, so I'll stop here. Our tackling with climate change would be great also. :D I liked this discussion, thank you!
Right! A few very powerful organizations, countries and people want to have access and use AI, for good and nefarious purposes. And then they want to prevent everyone else from using AI, because it's "too dangerous". I think it's time to look for other solutions, because this one is not acceptable.
The problem is capitalism, the profit maximization motive, fiduciary duty, etc; the solution, therefore, is obvious and inevitable. AI is not the problem, who owns it is. Automation is not the problem, the structure of our economy is. Electoral democracy is not the problem -- the lack of economic democracy, workplace democracy, is the problem.
@@RosscoAW Dude.... Socialism has already failed in USSR
@@anshuraj4277 And multiple times before that. AI's gonna be an issue in capitalistic hands or any other.
@@PolPot-ef1qq well.... U need capital to develop AI
Chat gpt "AI can have both positive and negative impacts on democracy and republics. It can improve efficiency and inclusivity, but also raise concerns about privacy and concentration of power. Proper regulation and ethical considerations are crucial to ensure AI benefits democracy."
Dictatorship, either.
Actually the missing ingredient is kinda obvious in my opinion. I can't believe they haven't already figured it out. Time stamp 5:50.
People should read "The Prefect" by Alistair Reynolds. Great insight on how AI will manage our democratic systems and what could happen if things go wrong. Very entertaining read as well if you into future history
I'm more worried about the unregulated internet and social media ruining democracy which is already happening. Nobody is talking about this
AI loves dictatorship, dictator loves AI too😅
(2014 Article which may be of interest): "Our 10 Favorite Killer A.I.'s in Movies"
My favorite killer ai movies are terminator series, I robot, and megan
All creators of technologies that may end up harmful to humanity gain a conscience! Especially in their later years and after they've helped release the genie from its bottle!
curiosity killed the cat
Also a.i. has appeal. If I were granted the access to turn on the following algorithms;H.E.M.I., Day1, Global Tristate, Education Revival and ultimately Old Earth Army as well. These all together create a New America. Should you see your ways forward ; that is to say.
I totally disagree. If AI humanoids were to replace politicians and leaders both on government and on local authorities levels, there would be no corruption, and a far better democracy. I think AI will be the only true democracy, and human leaders should finally disappear.
This is a very weak argument. We can clearly see with internet algorithms that AI doesn't promote democracy and freedom of speech but echo chambers. There would probably be no corruption left, but a lack of corruption doesn't mean a better world if the AI doesn't understand basic ethical principes or human emotional needs. Also another problem : if the AI seeks self-preservation, how likely is it to accept criticisms regarding its own algorithmic decisions? Especially if those algorithmic decisions are based on maximizing logic and efficiency at the detriment of human needs. At one point the AI may also come to the logical conclusion that humans are no longer useful and deserve to be eliminated not even because they are a threat to the AI but only for efficiency reasons.
0:00: 🤖 AI has advanced rapidly in recent years, with large-scale models like chatGPT showing impressive capabilities. 5:46: ⚠ AI researcher raises alarm about the dangers of AI and the misalignment problem. 11:21: ⚠ AI has the potential to be used for nefarious purposes, such as designing dangerous viruses and influencing elections, and it is crucial to have safeguards in place to prevent misuse. 16:40: 🔒 The immediate danger of AI systems is malicious users, so access to dangerous technology needs to be controlled through licensing and regulation. 21:56: 🔒 The speaker emphasizes the need for countermeasures to protect against dangerous AI with malicious goals, regardless of the timeline for its development. Recap by Tammy AI
That broke my heart. I just want to hug him and tell him he’s wonderfully made and precious. I pray that God has his hand on him and deep inside his heart. Lets agree to hold him and his family up before the Lord and ask him to protect his heart from this hideous pain. Let the Holy Spirit grow in him the courage of a Lion and the gentle kindness of the Lamb showing not self doubt but rather an abundant ability to forgive the wickedness thrown at him but let not take a single thing said out of hate from of their verbal venom be allowed to seep into his mind or heart. Let him have a gift of turning their words against themselves by not being effected at all but instead repaying their evil with Good delivered with confidence self love self expectance and knowing his true undeniable worth a a child of the Great I Am. Bless him in all that He is and all that he will be Amen.
😮😮😮👌👌👌👌
AI cannot operate without programming installed by engineers. The engineers can be evil to manipulate the users. It's not AI is dangerous, but the mind and purpose of engineers programming.
Like in terminator 3, ai can be self aware and infect every computer in the world and start a nuclear holocaust. Sophia is programmed to be a good robot. Watch out for angry robot like ameca.
The external world is offensive and artificial intelligence does not assist much at all. So there.
With this mentality, the poor people will pay, again, because there are chances that there will be some bad people in the lot. To deny people full access would be controlling as if the people were children. It is wrong. I say let the thing grow!
Don’t come to Canada dude
Wut
why would I want to live in a democracy. that's 2 wolves and one sheep sitting at the dinner table voting on what they want for dinner that night. a constitutional republic made up of laws protecting individual liberty is MUCH BETTER!!