W1siziisimnvbxbpbgvkx3rozw1lx2fzc2v0cy9uagugqnjpzgdlieluic9wbmcvymfubmvylwrlzmf1bhqucg5nil1d

Blogs

Bridge Bytes: Is this the end of AI as we know it?

16 Jan 10:00 by Harry Robinson

W1siziisijiwmjavmdevmtyvmtavntuvndkvnda4l25lb25icmfuzc01ywzlbnhuterqcy11bnnwbgfzac5qcgcixsxbinailcj0ahvtyiisijgwmhg0ntajil1d

Before we get started, there really is no need to panic. AI isn’t going anywhere. 

After significant advancements in 2019, ranging from automated delivery robots to litter boxes with machine learning, Artificial Intelligence is here to stay. The main worry today is that AI might be slowing in progress, not disappearing entirely. After a recent report by the BBC, tech experts have predicted a drop in public interest for AI. The hype is dying down, and that can have some serious effects on the development of upcoming products.

 

Words from the Godfather

The previous decade has seen an incredible increase in AI’s impact on society. In 2010, AI was present in more sci-fi films than actual products, but with technological progress we can now appreciate the weird and wonderful advances it has brought. Last week’s CES conference was a clear indication of just how many new releases incorporate AI, and the pinnacle of weird was none other than the £252 litter box called the iKuddle.

Yet there are some serious doubts for 2020 and the decade to come. Katya Hofmann, a principal researcher at Microsoft Research told the BBC “I have the sense that AI is transitioning to a new place.” And Katya is not alone in her hypotheses, as Yoshua Bengio made it clear that the work put into AI over the last decade will not be matched for some time. Yoshua, commonly referred to as the “Godfather of AI” alongside Geoffrey Hinton and Yann LeCun, is a professor at the Department of Computer Science and Operations at the University of Montreal. He has received numerous awards on his contributions towards the development of Artificial Intelligence and sits on the board of numerous research organisations, including one he founded himself ‘Element AI’. It’s fair to say that Yoshua wrote the book on AI, mainly because he did in 2016, so for him to cast doubts on the short-term future of AI is a significant red flag for the years to come.

 

The original G

So what’s actually behind all this doom and gloom? Why have professors and researchers turned their backs on AI? In short, it’s because of AGI. Yes, another acronym.

AGI stands for ‘Artificial General Intelligence’, and whilst it may seem like they’ve only added a G, the science involved is light years ahead. This leap was promised as early as 2014, as Google’s DeepMind thrust themselves into the spotlight with their £400m acquisition. Since then, many have jumped onto the AGI bandwagon. It wouldn’t be a discussion on tech if Elon Musk wasn’t trying to profit, and that’s just what he’s done with his new $1bn research company ‘OpenAI’.

This initial gold rush has resulted in very little progress with AGI development and the experts like Yoshua Bengio have come to the realisation that current techniques will only get us so far. AI has hit its ceiling, for now, and some serious innovations need to happen before it makes the leap to that extra G.

 

Always look on the bright side of AI

There’s been a lot of negative news on Artificial Intelligence in 2019, so let’s kick 2020 off on a positive note. Yes, there may be less advancement in the next couple years, but to label that as the end of AI is a discredit to the progress we’ve made so far. What do Netflix, Google and Amazon have in common? They all depend on AI in some form. If there’s anything we can take from this decline in hype for AI, it’s that we’ve already been blessed by the technological advancements of this last decade.