So, can ChatGPT do your job better than you?

I typed the question — “Explain what ChatGPT is Cockney style” — and within seconds the computer-generated response gushed out: “To ‘Ello gov!’ Let me tell you about ChatGPT, the AI ​​language model that can talk and walk. Look, ChatGPT is a powerful machine trained to understand and create human language.

“But be careful, because like any powerful tool, ChatGPT can be used for good or bad. So use it wisely and fairly, or you may find yourself walking on plank!

This is one that convincingly started Danny Dyer but ended up more Jack Sparrow, I’m sure you’ll agree.

While it’s not perfect, you can understand why ChatGPT has sparked so much controversy. You can ask him to write a sonnet about Sadiq Khan (“So let’s honor the noble son of London who worked to make his city shine…”) or a one-line summary of Prince Harry’s new book (“The raw account of his personal and mental health struggles …”).

There are even reports that ChatGPT-generated responses will pass the prestigious Wharton MBA course entrance exam in the USA. It’s a level of complexity that worries many but excites others, including Microsoft, which on Tuesday announced a multibillion-dollar investment deal with OpenAI, the company that created ChatGPT, to try to get ahead of its competitors in the AI ​​arms race. . Counting Elon Musk among its early investors, OpenAI is currently valued at around $30 billion – quite a feat for a company with just 350 employees.

ChatGPT uses a “deep learning model”, basically it examines a large amount of text on the internet and uses that data to figure out how to construct human-sounding sentences.

Meta’s chief AI scientist claimed that ChatGPT is not “particularly innovative” because its core technology isn’t too different from technology developed at other companies.

Whether that’s true or not, there’s no denying that ChatGPT is one of the most tangible uses of artificial intelligence ever – the term previously used for software that can do tasks that only humans could do (like writing complex prose) and potentially outwit us (no one’s latest as in chess, where it cannot beat artificial intelligence machines).

the story continues

At the same time as ChatGPT, we’ve seen the emergence of AI software like DALL-E (again from OpenAI) and Midjourney that can produce amazing images and works of art in seconds – the kind of stuff only humans could create until recently. . I’ve been writing for years about AI’s potential impact on jobs and society, and London’s position at the forefront of this change, thanks to British companies like DeepMind, Faculty and Stability AI.

Just as automated factories are reshaping the labor market by replacing many “blue collar” jobs with robots, some experts predict artificial intelligence will replace many “white collar” roles. The Bank of England thinks that up to 15 million jobs could be lost in the UK; Administrative and bureaucratic roles are among those most at risk, as they can most easily be replaced with advanced software. Other potential disadvantages are also easy to consider, such as artificial intelligence-generated fake news or students cheating by printing their assignments on ChatGPT. But that’s why I think the future of artificial intelligence may be brighter than pessimists think. For many employees, AI software will be an indispensable tool that helps us in the end.

Take a look at how Microsoft has included ChatGPT in Word and PowerPoint – which means your computer can increasingly do the tedious work of writing and putting together presentations as long as you point it in the right direction.

Google wants to do the same with Gmail so it can write emails for you – and they’re apparently planning a chatbot for search engines to make it easier to find information online.

Another good example is teaching where technicians conduct experiments to see how ChatGPT can be used to mark students’ work. If these approaches come to fruition, they can help free people from the most boring and time-consuming parts of their work and give us more room to focus on the things that are meaningful and important.

But the main reason we’re positive is that artificial intelligence can unlock innovations that we wouldn’t have found on our own. Perhaps the best case study is in London, where DeepMind researchers made surprising breakthroughs in unraveling how proteins, the building blocks of life, fold themselves in different ways. If this problem can be resolved, it could lead to new treatments for diseases such as Alzheimer’s and Parkinson’s, potentially enabling scientists to develop new sources of clean energy.

DeepMind’s software extracts large amounts of data and detects patterns much faster than a human could — but this raw AI computing power is also combined with the brainpower of humans who use the software’s findings to develop new insights and breakthroughs. Demis Hassabis, co-founder of DeepMind, says that using AI in this way “enables us to be much more creative, often trying more ideas in parallel.” Instead of replacing scientists, he supplemented his skills.

We’ve already seen ChatGPT listed as a co-author in academic papers – physicist Kai Polsterer describes AI systems as “hardworking assistants” who can free you to “do the cool, interesting science on your own.”

What we’re starting to find in science may be a better prediction of the future than a vision of a world where robots are destroying all our jobs. A work life that has the potential to be more interesting and at the same time more creative than we could have imagined.

I leave the final word to ChatGPT, who, when asked about the years ahead, replied: “AI can be both good and bad, depending on how it is designed, developed, and used.”

It sounds like a questionable politician’s answer – but it’s probably close to the truth.