Venture capitalist Marc Andreessen says he sees several advantages to a world dominated by AI coding agents and bots.
“The bots never get frustrated with you,” Andreessen told Joe Rogan during a recent episode of Rogan’s “Joe Rogan Experience” podcast, adding amid laughter that a bot “never gets drunk, never gets sick, never gets high” or files “HR complaints.”
Andreessen, cofounder of Andreessen Horowitz, said that right now in Silicon Valley, the “state of the art” setup is to run roughly 20 bots at once. That way, the agents can run concurrently, and the user doesn’t have to wait constantly to provide feedback.
“This is why people can’t go to sleep is because you’ve got 20 AI bots that are all as good as the best programmer in the world that are doing exactly what you tell them to do on every project you’ve ever wanted to do,” he said. “And they’re running 24/7, and the only thing you have to do is be there every 10 minutes to be able to give them feedback on what they’re doing.”
Another huge benefit, Andreessen said, is that bots don’t get emotional about work. He described a scenario in which a human employee is angry after spending two weeks on a project, only to be told that the final result isn’t right and needs to be changed. After changing the project, the employee is then told that the first result was actually better, and they have to change it back.
“The guy gets pissed at you because he’s like, ‘I just wasted my time,'” Andreessen said. “The bot’s like, ‘No problem, no sweat, whatever you want, and we can try it 12 more times if you want.'”
The reality Andreessen described illustrates the extent to which advancements in AI coding tools, including Anthropic’s Claude Code and OpenAI’s Codex, have fundamentally changed not only software engineers’ jobs but also their very existence, leading to workers keeping odd hours and open laptops as they try to keep up.
Nvidia CEO Jensen Huang said during the chipmaker’s recent quarterly earnings call that he foresees a world with “billions of” AI agents.
“I mean, we’re going to grow into it. But it will have billions of agents and those billions of agents will all use tools,” Huang said. “And those tools are going to be like PCs.” Just like us humans using PCs today, in the future, you’ll have an agent using PC.”
This “army of bots,” Andreessen said, will start popping up in programming, but it won’t end there.
“It’s going to start with coders, but then it’s going to be every other job,” he said. “So it’s going to be every writer, you’re already doing it, every writer is going to have it, every lawyer is going to have it, every doctor is going to have it.”
Andreessen said the next step is a world where each agent or bot has subagents or bots of their own, mimicking the management structure for human employees.
“This is just starting right now, but like when we’re sitting here, here in a year, I think it’s going to be routine to have 10 to 20 bots each that have 10 to 20 bots,” he said.
