Ever since the rise of artificial intelligence, there has been a raging debate on if and when the new technology can take over jobs that humans currently do. Various projections and forecasts have attempted to quantify AI’s impact on the workforce, but there is still no broad consensus on the subject.
However, we now have a perspective from Olivier Godement, head of product for business products at OpenAI, who highlighted three areas where AI is on the cusp of automation. Godement shared his predictions on a recent episode of the “Unsupervised Learning” podcast.
Which jobs could be taken away by AI?
1) Coding
Godement noted that automation hasn’t yet reached a stage where it can “essentially automate any white-collar job,” but added that we are “starting to see some quite strong automation use cases” in a few specific fields.
He said coding is one of the most obvious candidates for AI automation, pointing out that AI coding tools have become so integral that taking them away from engineers “would likely cause a riot.”
“The automation is probably not yet at the level of completely automating the job of a software engineer, but I think we have a line of sight to essentially get there,” he said.
2) Customer Support
The executive added that another area with a “strong case” for automation is sales and customer support. Pointing to OpenAI’s work with telecom giant T-Mobile, Godement said AI models are already producing “fairly good results in terms of quality at a meaningful scale.”
“My sense is we’ll probably be surprised in the next year or two by the amount of tasks that can be automated reliably,” he added.
Godement also pointed out two key advantages AI has over humans: infinite patience and multilingual capability.
“The model is infinitely patient. For people who need more time to get their answer, the model can take five minutes,” he said.
“There’s a long tail of languages, and if you have to staff customer support agents for every language, you’re not going to make it.”
3) Life Sciences and Pharma
Godement said his “bet is often on life sciences and pharma companies” being automated with AI, particularly in drug discovery. Referring to work with pharmaceutical firms like Amgen, he explained that drug development has two major components: research and experimentation, and administrative processes — the latter being extremely time-consuming.
“AI models are pretty good at aggregating and consolidating tons of structured and unstructured data, spotting diffs and changes in documents,” he said.
“Once we figure out how to properly version, audit, and release the model with the right permissions internally, we’ll see a bunch more adoption in life sciences — and the outcome will probably be more medication, new drugs essentially for people, which is pretty cool.”
About the Author
