Will Terrifying AI Destroy Humankind?

Unveiled

Watch Will Terrifying AI Destroy Humankind? - Unveiled on Dailymotion

Transcript

00:00In 1997, IBM's Deep Blue defeated the world champion, Garry Kasparov, at chess.

00:07In 2016, DeepMind's AlphaGo beat Lee Sedol, one of the world's top Go players.

00:15In the future, it will be much more than just board games.

00:19So what do you predict for tomorrow's AI?

00:24It was actually during ancient times that we first created myths surrounding artificial

00:28intelligence.

00:30Greek legends were the earliest, such as the story of Talos, a bronze automaton gifted

00:35with intelligence tasked to defend Crete.

00:38For most of our history, though, machines that could think have been far from the realms

00:41of reality.

00:42It wasn't until around the middle of the 20th century that serious research into AI

00:47began.

00:48Alan Turing was the earliest pioneer of the field, publishing Computing Machinery and

00:53Intelligence in 1950.

00:55In it, he coined the Turing Test, which is a way to determine if a computer can indeed

01:00think intelligently.

01:01It's been over 70 years since this paper, and now a variety of different AI is commonly

01:06used in daily life.

01:08Language translation, facial recognition, online advertising, search engines… there's

01:13an ever-growing list.

01:16Slowly it's starting to feel like there isn't a field we haven't applied it to.

01:19And clearly, we have quickly found a lot of benefits to AI.

01:23For example, they can easily handle large quantities of data, are free from human error,

01:28and of course, they can work 24-7.

01:30Generally, it seems that technology is improving society, and should continue to do so.

01:36AIs are a route to ultimate efficiency.

01:38However, many believe that we should be extremely cautious.

01:42Renowned physicist Professor Stephen Hawking once said, quote, the development of full

01:47artificial intelligence could spell the end of the human race, end quote.

01:51He followed this up by saying that AI would far exceed humans, since our advancements

01:56are limited by biological evolution.

01:59The late scientist thought we should be extremely careful.

02:02How justified, exactly, were his fears?

02:05Where AI is in its current state, already people worry it can be used maliciously.

02:10We're currently in a worldwide race for legal autonomous weapons, or LAWS.

02:15In 2024, the US Department of Defense promised $1 billion for the Replicator program, which

02:22aims to field thousands of autonomous war drones.

02:25Shortly, it's expected that drones will be able to use widespread facial recognition

02:30to target and attack specific individuals.

02:33Overall, warfare is one of the simplest fields we can apply AI to.

02:37And as a result, the United Nations has been debating a worldwide ban on autonomous weapons.

02:43Unfortunately, quite a few countries oppose such a ban.

02:46So, governments worldwide are weaponising artificial intelligence… but can we be sure

02:51that those AI weapons will stay loyal to their creators?

02:55While the US, for example, is against banning them, they still want to ensure humans are

03:00ultimately controlling them.

03:02The concept of emergent behaviour, however, implies we might not be able to control them

03:07forever.

03:08In the case of LAWS, it's when, in a near-future time, they're connected in such a way that

03:12they can easily communicate with each other, independent of human behaviour.

03:17Some military minds have pitched the idea for teams of hundreds of LAWS, for instance,

03:22connected in a widespread hive mind.

03:24No humans necessary.

03:27Communication between them would expand, but to where is almost impossible to predict.

03:32At the least, we might expect whole new tactics arriving via emergent behaviour, turning weaponised

03:37AI into a force of its own to be reckoned with.

03:40The Pentagon is reportedly making a big push to develop what some have labelled Slaughterbot

03:45Swarms… all of which means that, like it or not, they are likely to become a reality.

03:50Again, we have no idea what a heavily-armed and highly-connected flock of AI intelligence

03:56will do.

03:57Hopefully, fail-safes will be put in place to prevent serious issues… however, some

04:02worry that any attempted block or limiter will eventually be overridden.

04:06Meanwhile, the US Air Force is also working on something called Project Venom.

04:11This aims to develop powerful F-16 fighters which are capable of flying themselves.

04:16Currently, about $50 million has been invested… and on the bright side, AI jets will certainly

04:21reduce the need to risk human pilots.

04:24If we can rely on AI's loyalty, and if we have the proper fail-safes in place, then

04:29they should only ever be a danger to enemy targets.

04:32However, once again, this is entirely new ground.

04:36Can Venom really be realised exactly as its developers want it?

04:40Won't there always be a risk of it turning against its maker?

04:43Or of it misinterpreting or refusing mission orders?

04:47Or just going on a rampage?

04:49These are the sorts of huge questions that dog any plans to push forward.

04:53In general, as we haven't yet completely solved AI even in a non-military context,

04:59many believe it's just too far too soon to try weaponising it.

05:02One positive note is that most major nations agree AI should never be given access to nuclear

05:08weapons.

05:09Although, alarmingly, not everyone is quite in agreement here, either.

05:13But of course, AI isn't only about lethal autonomous weapons.

05:17Yes, they could prove our doom… but what about everything else in the AI bracket?

05:21Currently, AI still isn't truly sentient.

05:25We've likely all used some type of virtual assistant, such as Apple's Siri… but these

05:29are not intelligent enough to overthrow humanity, no matter how spooky they can sometimes seem.

05:35Truly sentient AI is predicted in the very near future, though… so should we be worried

05:39about that?

05:40In even a non-military setting?

05:43In 2017, news broke that Facebook had developed two chatbots tasked to converse with each

05:48other over a fictional trade negotiation.

05:51Machine learning was used to create them, and the chat was monitored.

05:55Scarily, the two bots quickly deviated from any predicted script.

06:00They developed their own language, and started conversing in this instead.

06:04It was at this point that Facebook shut the study down.

06:07Other AIs have done similar things, with Google's translation AI also creating artificial languages

06:13before now.

06:14Broadly, it's thought that the seeming nonsense to us acts as an intermediary language, an

06:19unreadable link to the machines.

06:22While it's true that neither of these examples is particularly dangerous, given their lack

06:26of power, both cases do highlight how an AI world could lead in all new and unknown directions.

06:33The Facebook and Google stories might easily be explained away as glitches right now…

06:37but what happens when there are more than one-off peculiarities?

06:41Are these small moments a sign of more significant things to come?

06:45Creative AI, the most readily available form at present, can generate images, videos, text,

06:51and solve equations.

06:52And the tech can already do a lot of damage… for example, by replacing certain jobs.

06:58Such as in China, where reports claim that about 70% of video game illustrators have

07:03been axed, partly due to growing reliance on AI.

07:07Gamers have critiqued the AI products, saying they lack human creativity… but there's

07:11little sign of the trend stopping.

07:13Perhaps the most infamous contemporary issue of all, however, is the ongoing appearance

07:17of deepfakes.

07:19These involve using someone's likeness to create an exact copy of that person… which

07:24rapidly leads to false images and video involving them, made without their consent.

07:29Such AI was the trigger for the Hollywood Strikes of 2023… but these tools could ultimately

07:34impact far more than simply those in the media.

07:37How far can these new realities take us?

07:40On the one hand, as current, generative AIs are usually trained using human data, it's

07:46at least thought and hoped that they couldn't yet learn to do things that we can't.

07:51Generative results draw on what's been done prior.

07:53In other words, human knowledge will likely be the limiting factor.

07:56And so, the general consensus is that AI in this state can't directly turn against us.

08:03Replace us?

08:04Maybe.

08:05Dislike us?

08:06Possibly.

08:07But break away from us?

08:08Probably not.

08:09There is a darker extension, though.

08:11True AI, also known as Artificial Superintelligence, is a different ballgame.

08:17Right now, it remains in the realm of science fiction.

08:20But were fiction ever to become fact, then it's proposed that this more advanced AI

08:25will learn so well that it will exceed human understanding.

08:29Humans will no longer be the most intelligent lifeforms on Earth, and we'll all be painfully

08:34aware of our demise.

08:35Today, many may consider it speculative territory, but others say that it's vitally important

08:40for us to speculate in order to heat it off.

08:44According to Stephen Hawking, for one, there are no physical laws preventing AI from one

08:48day – perhaps one day soon – operating better than the human brain can.

08:53At which point, it could turn against us, just as easily as it could do any number of

08:58other things.

08:59Will supreme AI crave power?

09:02Will it recognise life on Earth as valuable?

09:05Or see it as a threat?

09:06Or merely an annoyance?

09:08Will it one day scrub back through all the media of now – the books and films and YouTube

09:13videos – with admiration, relish, or disdain?

09:17For all the concerns, generative AI should remain beneath us, but is generative only

09:23the first generation?

09:24And will what's coming next care two hoots about us that came before?

09:29What do you think?

09:30Is there anything we missed?

09:31Let us know in the comments.

09:33Check out these other clips from Unveiled, and make sure you subscribe and ring the bell

09:37for our latest content.

Category

Transcript

Recommended