AI Is Definitely Coming for Your Job

One of the big promises of AI is that it will automate repetitive tasks, freeing up humans to focus on more creative work. But the “freeing up humans” ideal only works if we still have jobs that employers value enough to pay for. 

I bring this up because when it comes to concerns that gen AI will snatch jobs from people, various kinds of writers land at the top of the list. That’s because AI tools can take a whack at composing everything from emails and resumes to novels and wedding vows and deliver text “for you to review in seconds,” as Grammarly has noted. 

I don’t buy that AI will lead to the demise of writing jobs, let alone that it will replace writers. Neither do others, with author Ted Chiang questioning whether an AI can produce “art,” and novelist Anita Felicelli arguing that “an algorithm trained on well-known books can’t find what’s both moving and surprising the way” skilled authors can. But tools like ChatGPT “will figure out three-act structure, cliffhanger endings for chapters and what events will titillate readers faster to maintain reader attention.”

But I do believe — along with pretty much everyone else who studies the future of jobs — that gen AI will reduce opportunities for all kinds of workers, writers among them. All kinds of jobs will be outsourced to the robots as companies across industries look to cut costs and get a return from their investments in AI.   

I mention this because I’m hearing more and more among my circle in Silicon Valley that while we writers should be concerned, a group as likely to be affected by the AI displacement effect will be those who helped create AI in the first place: software engineers. 

Of course, you’ll find plenty of arguments about why AI can’t replace engineers. But that doesn’t mean software creatives will be any less vulnerable than writers to a rethink of their role, as tech companies turn to AI tools to boost productivity among a group that’s typically been their most expensive workers.  

My musing on this is prompted by recent comments from two high-profile tech executives. 

First, the chief of Amazon Web Services, Matt Garman, predicted that the company’s software engineers “could soon be forced to find responsibilities other than coding due to the rise of AI,”  the New York Post reported in August, citing leaked audio of his remarks to employees obtained by Business Insider. “It just means that each of us has to get more in tune with what our customers need and what the actual end thing is that we’re going to try to go build, because that’s going to be more and more of what the work is as opposed to sitting down and actually writing code,” Garman said.

And last week, the co-founder of Anthropic, Daniela Amodei, told The Information about how the maker of Claude is using the chatbot to help its developers write code. Though Claude “can’t fully replace engineers and needs ‘some coaching,'” the publication reported, quoting Amodei, Anthropic developers have “dramatically increased their productivity.” That’s great, the publication continued, as Anthropic attempts to address the “overriding question for artificial intelligence developers [as to whether] they can generate enough revenue to offset the high costs of compute talent.”

To be sure, neither Garman nor Amodei predicted the demise of software engineers, just that AI will change what they do. When I do a Google search on how artificial intelligence will alter software work, Google’s AI Overview feature tells me that AI can “automate repetitive tasks like code reviews, testing, and debugging, allowing developers to focus on more creative work.”  

That’s all well and good — assuming they still have a job that employers value enough to pay for. 

If you’re interested, Bain & Company shared thoughts last month on how software development will change in an AI-powered world. Meanwhile, Fast Company reported that the founding president of Google China, Kai-Fu Lee, “agrees there will be countless new jobs created by the effort to optimize AI, which will require a human touch.” What jobs? “Positions like Prompt Manager, AI Trainer, AI Auditor, AI Ethicists, and Machine Managers will be necessary in order to help companies develop and use AI in a safe, responsible, and effective way,” the publication said. 

What he seems to be saying is that if you’re concerned about AI coming for your job, you might want to upskill to a job centered on the care and feeding of AI. I don’t know about that. But I do know that if software engineers need to start rethinking what they do, then it’s probably time we all reflect on how AI will change our jobs in the not-too-distant future.  

Here are the other doings in AI worth your attention.

Why is Mark Zuckerberg leg-pressing chicken nuggets?

How big a deal will AI be when it comes to allowing you to create high-definition or photorealistic video from simple text prompts? 

A pretty big deal, based on the gasps heard from Hollywood when OpenAI demonstrated its photorealistic text-to-video creator Sora back in February (and said the tool will be made available later this year). If you don’t remember, or know, anything about Sora, I encourage you to watch this YouTube demo reel and check out Joanna Stern’s assessment in The Wall Street Journal. “Welcome to the next ‘holy cow’ moment in AI, where your words transform into smooth, highly realistic, detailed video,” Stern wrote. “So long, reality! Thanks for all the good times.”

Well, there were some more gasps when Meta announced this month that it was moving into the text-to-video space with a tool called Movie Gen. The tool promises to give everyone from aspiring filmmakers in Hollywood to creators who just like making videos for their audiences a way “to use simple text inputs to produce custom videos and sounds, edit existing videos, and transform your personal image into a unique video.” 

When Movie Gen is released to the public for use on Instagram (Meta didn’t say when exactly), you’ll be able to create full-motion video up to 16 seconds long from a single prompt, or from an uploaded photo of someone. Or, as Ars Technica described it, Movie Gen is an AI system that can generate “deepfake videos from a single photo.”

Concerns like that are why Meta said in its blog post that it will “work closely with filmmakers and creators to integrate their feedback.”

Check out Movie Gen’s AI-generated hippo video, and a few other demos, here. Or you can watch a Movie Gen version of Meta CEO Mark Zuckerberg leg-pressing chicken nuggets, on his Instagram account, here (at least I think they’re chicken nuggets, against a backdrop of fries). Warning: Once you watch it, you can’t unsee it.  

Godfather of AI — and unapologetic doomer — wins Nobel Prize 

Former Google researcher Geoffrey Hinton, described as one of the godfathers of AI, shared the Nobel Prize in physics for his work in using “tools from physics to develop methods that are the foundation of today’s powerful machine learning,” the Royal Swedish Academy of Sciences, which awarded the prize, wrote in a release. “Geoffrey Hinton invented a method that can autonomously find properties in data, and so perform tasks such as identifying specific elements in pictures.” 

What’s notable about the win is that Hinton has become one of the most vocal critics of AI, “warning the machines could take over the world,” The Wall Street Journal noted. “Hinton’s Nobel win has provided a new platform for his doomsday warnings at the same time it celebrates his critical role in advancing the technologies fueling them. Hinton has argued that advanced AI systems are capable of understanding their outputs, a controversial view in research circles.”

Hinton told the WSJ that the Nobel Prize win hopefully “will make me more credible when I say these things really do understand what they’re saying.”

Hinton wasn’t the only Googler recognized for his contribution to AI. Demis Hassabis, co-founder of Google’s DeepMind AI division, and his colleague John Jumper were awarded the Nobel Prize in chemistry after developing “an AI model to solve a 50-year-old problem: predicting proteins’ complex structures,” the Royal Swedish Academy of Sciences said. 

Also worth knowing…

Google Search added new AI features that let you “ask your question in any way you want — whether you type a query, search with your camera or simply hum a tune,” the company said in a blog post. The news comes as the US Department of Justice said it may consider asking that the company be broken up, after a judge ruled it has an illegal monopoly in online search, Reuters reported. (The case is expected to go on for years, so legal experts don’t expect Google Search to change anytime soon.)

CNET’s Katie Collins recounts the challenges Apple is facing as it works to gain ground in the AI market. “The company promised Apple Intelligence would be here ‘in the fall’ and it looks like it will make good on that promise, but it arguably would’ve had a bigger impact if it had arrived ready to go on the new iPhone last month,” Collins says.

Apple researchers released a paper describing an AI research model called Depth Pro that VentureBeat reported is “able to generate detailed 3D depth maps from single 2D images in a fraction of a second — without relying on the camera data traditionally needed to make such predictions.” Why is that worth knowing? Because it could “significantly advance how machines perceive depth, potentially transforming industries ranging from augmented reality to autonomous vehicles.”

If you’re in Southern California, AI LA is running a series of events through Oct. 18, including a discussion around responsible AI development.