10 ms·
Yoshua Bengio: ‘The dangers of abuse are real’
- woliveirajr 8y ago> "has raised concerns about the possible risks from misuse of technology" Isn't misuse possible in all (or, at least, almost all) technology? Drugs can change humor, physical condition, hormones, and I think can be miused even as a long-term weapon over the population of some nation. Nuclear. Explosives. Social medias. TV programs. Data processing with punch cards. Drones. Knifes. Herbicides. Small plastic rejects. I understand the reasons Bengio tries to make, but it seems that it is the same problem with all technological stuff: someone, somewhere, will find a way to use it against others, and will cause minor or giant consequences.
- naringas 8y agoin my opinion the possibility of misuse a defining trait of technology.
- jononor 8y agoMost of your examples are covered by rules and regulation in order to try to get the most out of the technology without too much adverse effects. The point here is that we should do the same for "AI", and that we should start thinking about this now.
- novaRom 8y agoNot a secret that largest military companies are actively investing into AI. Enormous opportunities and dangers. And it's not just computer vision and robotics - almost every aspect of modern warfare will be AI-extended.
- caiob 8y ago> People from outside Canada are coming to Canada to do research in AI. This is true. As a resident of Montreal, I can attest to the fact that we've become THE hub when it comes to AI research and development. Specialists are moving up here en mass to do their thing.
- i_am_proteus 8y agoHow will regulating AI be any more successful than the 1990s attempts by the US to regulate cryptography?
- jagger27 8y agoThe accessibility story is nearly identical. So I'd say nope.
- oldjokes 8y agoIt's the same story in all new tech breakthroughs: the rich and powerful are already leveraging AI and the complete lack of regulation to geometrically explode their hoardes and power, everyone else not so much. Acting like this is some far off future event is absurd, people are already making trillions of dollars off of AI and personal data.
- monkeynotes 8y agoAI is like a mountain. No matter what humanity does, someone will summit just because it's there. We can control businesses to an extent, but a curious person in the basement, no one has control over that. And good luck holding the military back. One way or another humanity is on its way to the inevitable, and it's our evolutionary destiny because we are basically hard-wired to pursue the curiosity that is AI.
- everdev 8y agoI think there was once a yogi or guru-type that criticized the denuclearization effort saying why get rid of nukes if we still have the mindset that can justify using them? I can't help but think the same is true of AI or any technology that could be used for good or bad. We spend a lot of time trying to control the tools, but less time trying to improve culture and ethics of the operator.
- canjobear 8y agoWe don’t know how to improve people’s ethics.
- BucketSort 8y agoYes we do. You change the social goals. A materialist society will always bring us to the brink of destruction. Also, having poverty and fear loom over people is a terrible idea for the basis of a civil society.
- macawfish 8y agoWonderful perspective, thank you!
- heavenlyblue 8y ago>> You change the social goals. How do you make the majority of the society believe in the same social goals without resorting to tales a-la Christianity during the fall of Roman Empire?
- novaRom 8y agoAI arms race is real. Will outcomes be good for civilization like those of space race (Internet, Communication) nobody knows.
- peterwwillis 8y agoThey keep using the term 'AI' when they seem to mean 'software'. We have all these problems regardless of how 'intelligent' the artificial machine is.
- 0x8BADF00D 8y ago> You have expressed concern that corporations have ‘stolen’ talent from academia. Nobody is stealing anything. What they don’t realize is that the free market will incentivize AI researchers to go to the private sector. It’s just simple economics.
- webmaven 8y agoIf private industry are using inflated (in the short term) valuations and low tax rates to fund a talent war, but publicly funded educational institutions can't compete by increasing funding (through governments raising taxes or printing money), then the playing field isn't exactly level, is it?
- wskinner 8y agoUniversities are nonprofits and pay little / no taxes.
- webmaven 8y agoYou missed the point: Raising government taxes increases the funding available for those institutions.
- GuiA 8y agoPublicly funded educational institutions can compete by reassigning some of their budget dedicated to sports and administrative roles to their researchers, for starters. For instance, the public US university I attended for grad school has an endowment of about $1B, and spent ~$150M on athletics in the past year. A post doc salary there is about $40k a year. Why would anyone take this over $200k a year (with lots of sweet perks) at eg Google? Because they are passionate about academic research (and potentially teaching), something that the university is very aware of and has no problem using to their advantage. Talking about level playing fields here seems disingenuous.
- webmaven 8y ago> Publicly funded educational institutions can compete by reassigning some of their budget dedicated to sports Athletics programs are a huge driver of both alumni donations and student enrollment. Unfortunately. Not to mention licensing revenue from apparel and the like. > Because they are passionate about academic research (and potentially teaching), something that the university is very aware of and has no problem using to their advantage. This was more effective when the pay disparity was smaller.
- alwaysanagenda 8y agoBengio is asked: What will be the next big thing in AI? > "Deep learning, as it is now, has made huge progress in perception, but it hasn’t delivered yet on systems that can discover high-level representations — the kind of concepts we use in language. Humans are able to use those high-level concepts to generalize in powerful ways. That’s something that even babies can do, but machine learning is very bad at." I read this as: "We have super-advanced skip-logic software that can produce specific results when provided a large enough data set, but "intelligence" as it is defined, does not exist." AI is really just sophisticated software algorithms. In my opinion, there is no true artificial intelligence, and it will be unlikely that we will ever create such a thing for quite some time, if at all. AI is being used as a buzzword to garner attention. It seems much more likely that we will build a brain-computer interface before true AI, and it will prove more efficient than what we have today, which is effectively many computers churning through a super-long list of "if-then" statements.
- xthestreams 8y agoThis joke about "if-then" statements has gone too far, and when people cite it it's clear that they don't understand the basics of how ML works today.
- drak0n1c 8y agoMeanwhile, Google has cancelled its AI ethics board in response to an internal employee petition complaining about a mainstream conservative being included in the 8-member committee. https://www.forbes.com/sites/jilliandonfro/2019/04/04/google-cancels-its-ai-ethics-board-less-than-two-weeks-after-launch-in-the-wake-of-employee-protest/ https://www.forbes.com/sites/jilliandonfro/2019/04/04/google...
- Dobbs 8y agoThe heritage think tank is a known anti lgbt group. They spend significant resources in the US and abroad to fight against lgbt rights. I for one and fking sick and tired of my existence being subject to debate due in part to these groups.
- satokema_work 8y agoWhat do LGBT issues have anything to do with AI discussions? I ask purely for information.
- Dobbs 8y agoAI is used to make decisions. It impacts every part of our lives. It encodes the biases of its creators into. How does that not affect LGBT folk? When AI starts being used t decide who to hire or if someone should be offered a rental then what happens when it is biased against LGBT folk?
- CamperBob2 8y agoIt means you can't assume the Heritage representative is going to participate in good faith. They are bringing a religious agenda to the table, which has (or at least should have) no place in the discussion.
- drak0n1c 8y agoHere's an article detailing their policy positions. No where do they oppose your existence: https://www.heritage.org/civil-society/commentary/dont-typecast-conservatives-they-too-respect-the-dignity-lgbt-persons https://www.heritage.org/civil-society/commentary/dont-typec...
- luxuryballs 8y agoIt always seems like FUD to me when I read these articles. What is the specific concern? Can we get some examples? Sometimes people are like oh no AI is gonna take over! And I just think about how hard it is for humans to integrate two software systems and laugh. What are we afraid of it doing? If it gets out of control can’t we just you know... cut the power?
- superhuzza 8y agoDid you even open the article? "Killer drones are a big concern....dangers of abuse, especially by authoritarian governments...AI can amplify discrimination and biases". Bengio is specifically talking about malicious actors using AI. So saying "just unplug it" isn't really applicable, especially when it comes to governments.
- luxuryballs 8y agoyou don’t need AI to make killer drones
- jhayward 8y agoThis is an example of 'whataboutism', and is used to derail a discussion without actually responding to the point.
- luxuryballs 8y agomy original point was already a response to your quote from the article, what is there to expound on? the problem isn’t an AI problem it’s a malicious actor problem
- auber 8y agoFUD is a really stupid term. If there is any fear or uncertainty or doubt present in someone’s remark, then they are “FUD?” You could say that about almost anything. “I don’t think it would be a good idea to ride our bikes into that forest pathway because there seem to be lots of trees to run into and it probably isn’t kept up so there may be potholes and generally the whole thing doesn’t sit well with me.” “That’s just FUD, I’m going in.” This is the classic childhood story where one kid was smart enough to not go in and the other kid ends up with a broken arm and a concussion. My point is that not everything can be mathematically proven to be bad. Sometimes you have to reason about it and you have to deal with uncertainty. You can’t solve every problem by trying it out. Some things are better to not try. And AGI is the best example of that maybe ever. With AGI the burden is on the proponents to prove beyond any doubt that AGI is completely safe. The burden is not on its detractors to prove that it will be bad for humanity. It’s only logical. If I am wrong and we abstain then we lose nothing. If you are wrong then we lose everything.
- aaron695 8y agoBored with AI 'experts' telling us it's dangerous. Unless they want to specifically explain out a danger, they add nothing that any kids who's seen The Terminator can do. > Killer drones are a big concern. There is a moral question, and a security question Other than your morality, which I don't care about, what specifically could happen? Be specific else it's pure FUD. NK could have introduced Ebola to the USA for decades, but they don't. So they, along with the rest of the countries, aren't going to 'just' introduce 'killer' drones.