Large Language Models & Generative AI: What Is Our End Game?

Large Language Models & Generative AI are awesome. But what is our end game as a human race?

It must have been when I was around 9 years old, in the early 80s: I watched Star Wars for the first time and was hooked up for life. In the meantime, I started writing some code (GW-Basic ??) on our MSX gaming computer until my older brother kicked me off to play Konami games. What I also loved doing. It was an epic decade!

When I stumbled upon bots in 2018, I felt truly blessed. I finally found my calling. My destiny. Enabling humans to have conversations with machines: what else could I wish for?!?

As a kid, I never dreamed to live long enough to witness where we stand today. In fact, until a couple of years ago I was still quite na?ve. I shared the fear small languages would disappear because the big tech was only focusing on big languages, which made perfect sense of course. I also thought it would take a decade before we would be able to chat with a bot about basically everything. In any language! About the world, life, the universe, and whatever rocks your boat.

I don’t want to spoil all the fun now, but please allow me to take us out of the #ChatGPT / #OpenAI / #BARD / #LLM bubble we are all in. Let’s zoom out for a moment, instead of zooming in even deeper.

Well, it's not as bad as BB-8 thinks, I just want to ask some fundamental questions:

1. Will AI make humans smarter or dumber?
2. Who will still have work in a couple of years?
3. Will AI make the gap between the rich and the poor even bigger?
4. Should tech giants take the lead in implementing a global basic income?
5. How should we, as an industry, deal with our increasing responsibilities?
6. What is our end game with all this automation in general?

1. Will AI make humans smarter or dumber?

This is obviously the first major concern teachers and parents had when they realized the power of ChatGPT. Meanwhile (most) students thought: “Cool, this is a huge time saver!” ??

Recent developments removed the complete fundament of our education systems. What is the use of learning how to do something, if a computer can do it faster and perhaps even better? Of course, it’s not only about the end product but about learning how to approach certain assignments and projects. The question remains however if schools and universities shouldn't teach way more about how to handle and validate all information that is now more easily available than ever before. Because the people who know how to deal with it, won’t lose their job.

2. Who will still have work in a couple of years?

AI will increase our productivity exponentially, and 2023 will be the year we will all remember. It's when the world really started to change. Rapidly!

What does it mean for jobs? Let’s take a digital agency for example, that produces (video) content for their clients. With speech-to-video software, already available today, they can say: “Please make me a 30-second promotional video, about why people should consider moving to Miami. In fact, I want to do some A/B testing, so make one version more focused on the different areas of the city, and another one focusing more on restaurants and nightlife”.

In the beginning, this business profits because they are able to handle way more projects at the same time than before. But at a certain point, clients will no longer accept their current pricing. In the end, as in almost any industry, fewer people will be doing more work with the help of software.

I tried to come up with a list of jobs that will not be impacted by AI and I didn't come very far. Therefore I asked ChatGPT for help:

3. Will AI make the gap between rich and poor even bigger?

It looks like AI will increase the gap between the rich and the poor. You and I don't have to worry, we are on the right track obviously. With businesses spending more and more on software and less on labor, the technocrats will populate the top of the pyramid, so to speak.

In the middle, there will probably be a huge group of people for whom there's not enough work available anymore. Current developments are sometimes compared with the introduction of the typewriter and later the computer. New jobs will emerge but with the vast majority of people in modern countries having office jobs, and exponentially increasing automation rates, the impact will be way bigger this time.

On the other hand, with software getting more advanced, work that's now only doable for people with a certain set of skills and sufficient brainpower (like conversation designers/AI trainers), will become within reach for more people.

If we indeed end up with a huge middle segment without work, boredom will be our biggest challenge. One of the main tasks of people in the lowest section of the pyramid would then be entertaining the people in the middle, and making sure the WiFi works.

On the other hand, we see technology improving wealth in for example the very poor regions in Africa. Thanks to tech, clean drinking water, electricity and WiFi becomes available to millions of people, meaning they now have the time and means to work more, start a business and get more education. This will unlock a huge new workforce for which wealth increases.

4. Should tech giants take the lead in implementing a Universal Basic Income?

I think they should and I think they will. Not sure when though. As most of you know Elon Musk is a big supporter of a Universal Basic Income, which will for sure help.

Link: https://www.businessinsider.com/elon-musk-universal-basic-income-physical-work-choice-2021-8

5. How should we, as an industry, deal with our increasing responsibilities?

That's a complex question to answer. How to make sure AI is unbiased is worth a separate article by itself. The recent suggestion to freeze all training of AI for a certain period has pros and cons obviously.

I do for sure think we need some sort of global committee, to make agreements on how to leverage AI, or at least warn in time for the impact. Something comparable is already in place to agree on how we deal with everything outside the Earth's atmosphere.

It's very complex though: will "the West" trust China for example, and vice versa? And how can we be sure technocrats like Elon are impartial? One of his companies might benefit from the suggested freeze. Besides that, how can we make sure companies really stop working on their AI? And do we want this kind of interference in our free economy?

6. What is our end game with all this automation in general?

To be honest: I really don't know, but would love to hear your thoughts! Let's discuss in the comments and/or during one of the #Botrepreneurs #BeersAndBots networking events ??