Is AI Eating Everything ?

I’m starting to see a lot of commentary that ‘AI is going to eat everything’

This phrase can likely be traced back to Marc Andreessen’s famous 2011 statement that “software is eating the world”.

He meant software companies were disrupting and dominating traditional industries through scalability, marginal costs, and digitization i.e. how Uber ate taxis, Netflix ate Blockbuster, etc.

The same logic is now being applied to AI, but with much higher stakes because AI isn’t just another tool layer, it’s a general-purpose intelligence amplifier that can potentially automate human knowledge work at massive scale.

So, is AI really going to eat everything?

It’s already eating huge chunks, but the pace and limits look different depending on which timescale and domain you zoom into.

Software / SaaS: AI agents and frontier models are starting to replace or deeply compress traditional enterprise software workflows. SaaS subscriptions can increasingly be replaced by asking an agent to create the subset of features you really care about.

Creative & media industries: GenAI has already swallowed large parts of stock imagery, basic copywriting, junior graphic design, simple video editing, and entry-level coding. The volume of AI-generated content online is now so massive that it’s creating model collapse / data pollution concerns.

White-collar knowledge work: Entry-level analysis, paralegal tasks, basic research, report drafting, some initial customer support tiers, marketing copy, and large parts of software engineering are being automated or massively AI augmented.

Scientific discovery & R&D: AI is accelerating strongly here. In these fields it really does feel like its slowly “eating” human-only processes.

So what jobs would I tell my kids to go get today if they were just starting out?

Physical-world tasks requiring dexterity such as an electrician, plumber, landscaping etc. Having a trade is becoming the new white collar profession.

Activities that are deeply social, emotional, status-driven, or tied to human authenticity such as live music performance, artisanal crafts marketed as “human-made”, high-value sales negotiation etc.

Would I advise computer science?

I’d advise they learn as much about using AI as possible as people who know the tools and how to get the best out of AI models will be those who find employment, but for computer science disciplines such as coding, if not yet fully eaten, they soon will be.

Its entirely possible AI will create new categories of work we can’t yet name, just as the internet created “social media manager”, “cloud architect.” etc.

AI is not a bubble, the “eating everything” commentary is partly hype to keep capital flowing, but unless you are heavily involved in the industry its a little frightening how unprepared and unaware most people are in terms of what is coming.

AI is like the predator in the famous movie Alien, that’s already consumed a few major species (SaaS layers, entry-level human labor, parts of creativity) and is eyeing other species. Whether it eventually eats everything depends mostly on how far scaling laws and algorithmic improvements pan out, along with energy accessibility (fusion, solar, etc.) and what pushback there is from regulation, and society at large.