What’s all this ‘chatbot’ about then?

OpenAI started life in 2015 as a research organisation with a mission to migrate AI safely into humanity. On the 30^th November 2022, they released a landmark AI chatbot, called ChatGPT, that has subsequently been doped AI’s ‘iPhone moment’. It was the fastest adopted internet application in history, growing to 100 million users just 2.5 months after its launch, much faster than previous records set by the likes of Instagram and TikTok.

Their success kicked off an arms race in AI investment, as after many decades of research, suddenly something useful had been produced. In today’s world, attention is one of the most valuable resources, so both big tech and little tech scrambled to understand what OpenAI had produced and looked to replicate and compete with their technology.

As business after business released their AI investment plans, parts of the tech market which had operated mostly in the shadows for 10 years, became really interesting. Nvidia is the poster child of the AI investment boom and in particular head up the so called magnificent 7 in the US stock market, namely Apple, Microsoft, Tesla, Meta, Amazon and Alphabet.

Despite a chunk of time passing, it is worth saying that as an individual consumer it remains difficult to see what all the hype is about. Most of us don’t yet use AI in our day to day lives, either personally or within our businesses. Outside of the better Chatbots which can help us write letters and summarise meetings, there isn’t that much available yet, so it is natural to question this market trend.

Most of the changes have happened below the surface. Much investment has so far been channelled into infrastructure, chips which provide power, data centres, energy sources and research on how to improve models. AI is a lot more computer intensive than traditional tasks, so for us to be able to use AI in our phones, they need to be significantly more powerful. That’s similarly the case with laptops, desktops, data storage… everything now needs to be bigger, which is a key reason why we haven’t seen and felt AI change our lives yet.

The second big challenge is that it needs to be better. Yes, it is interesting that ChatGPT can recite pretty much everything about everything, but two years on it still makes stuff up. AI doesn’t yet always tell the truth, in many ways telling you what you want to hear, rather than what is necessarily right. Trust is crucial in all industries, so those creating the models that will power future applications still have significant work to do to make things useable.

So, in short there’s a long way to go to get to the point where there are significant and tangible benefits. Added to that, these companies are now priced at eye watering multiples of earnings. They need a lot of things to go their way to achieve their year-on-year growth targets.

From a portfolio perspective, we have reduced our exposure since the start of this year, not just to tech stocks, though also to the wider US market. It’s not that we are downbeat, we are simply cautious and optimistic. Just like the internet revolution, it was very difficult to predict the winners from the losers, and there were significant bumps (as well as one very famous crash) along the way. We still have exposure to the technology sector of course, but we also think AI will provide broader benefits to all industries and are therefore focused on businesses that are leaders in their industries and are open minded to adoption when they can see the value. “Good things take time, as they should”.