The backlash against artificial intelligence feels distinct this time because it is no longer just about the technology itself, but about the economic conditions people are living in. A reader named Josh Vanderberg wrote in to suggest that the anger is a projection of financial hardship rather than a rejection of natural language interfaces. He argued that while modern computer systems are frustrating, people will eventually embrace them if they offer a better way to work. He noted that Canva’s CEO understands this problem and sees AI as the solution to the clunky menu systems that force users to map their creative thoughts into rigid hierarchies. The reader stated that people want to tell Photoshop what they want in words and forget the rest.
Other feedback echoed this sentiment. One person wrote that the issue is not automation, but the price of goods. They added that when AI provides cheap options, people will be happy. A third comment highlighted that the “software brain” argument misses three things regular people actually care about: control, agency, and meaning. The writer noted that people like doing things because that is where meaning comes from.
Origin of the rants
Host Nilay Patel explained that his direct-to-camera segments, often called rants, stem from his background as a writer. He stated that he feels he should make another riff on the “People do not yearn for automation” topic, delivering it as a full Dennis Miller-style rant. He said he would have written this as a column a decade ago, before podcasts were a primary format. He believes that writing a column no longer has the same impact as staring at a camera and speaking directly to the audience. He noted that his magazine always prioritised data getting to your head.
Patel said he chose to make the “software brain” video at that specific moment because the excitement around AI was so high. He saw the hype cycle failing to match the lack of verifiability in every domain except software. He explained that software engineering is verifiable because you can run it through a compiler to see if it works. While AI has become very good at making software, he argued that once you move outside of that domain, verifiability becomes much harder.
He pointed to a recent episode with reporter Robert Hart about AI and math as an example. He noted that there is some amount of verifiability in mathematics, where you can burn tokens and solve unsolved problems. However, he questioned if there is verifiability in generating novel drugs, noting that you must inject people with them to know if they work. He said there will be other domains in the next ten years that present similar challenges.
Patel stated he wanted to put a marker down to say that while the power of software has remade the world, viewing it as a universal framework will lead to problems. He explained that he wants his rants to hit and that they are all born from reporting. He said he feels he has the depth and foundation to write something like “software brain” and then send it to the people being discussed. He believes that conversation gets much bigger than him just ranting.
He admitted that episodes where he talks sometimes outperform episodes with CEOs. He called this a bad incentive loop because he wants everything to come from conversations he has had with people. He wants the audience to be able to check his work and see that the conversations are genuine and honest.
Producer Nick Statt noted that Patel’s thesis for “software brain” has borne out over the last few months. He said they received feedback from people who disagreed with some premises. He added that it was not that people were saying they were not unhappy with AI, but that they thought they were wrong about why they were unhappy.
When asked how he responds to these criticisms, Patel said he would take them in turn. He called Josh’s comment very thoughtful and said he did not disagree that all modern computer interfaces suck. He mentioned he has used Liquid Glass and refuses to upgrade his computer to macOS Tahoe. He stopped mid-sentence before finishing his thought on the upgrade.
What it means
For creators and makers, this feedback suggests that the conversation about AI needs to move beyond technical capabilities and address the human need for control and agency. The technology may offer efficiency, but if it removes the sense of doing things, it will face resistance regardless of how cheap or powerful it becomes.
