Has AI already rotted my brain?
Five years ago, I bought an e-bike. At the time, the motor-equipped two-wheelers were burdened with an iffy reputation. Was it way easier to get up a hill on one than on a bike without a battery? Absolutely. Did that mean people who rode them were lazy or even cheaters? Some cycling enthusiasts thought so.
But what if the boost provided by your e-bike motivated you to make longer trips and more of them—all powered, in part, by your own pedaling? Having logged almost 10,000 miles on my Gazelle, I'm certain it's been a guilt-free boon to my well-being. Data backs me up.
I thought about that recently while reading about a new study conducted at MIT's Media Lab. Researchers divided subjects ages 18 to 39 into three groups and had them write essays on topics drawn from the SAT questions answered by college applicants, such as 'Do works of art have the power to change people's lives?' One group relied entirely on unassisted brainpower to complete the essay. A second group could use a search engine. And the third could call on ChatGPT.
The study subjects wore EEG helmets that captured their brain activity as they worked. After analyzing that data, the researchers concluded that access to ChatGPT didn't just make composing an essay easier. It made it too easy, in ways that might negatively impact people's long-term ability to think for themselves. In some cases, the ChatGPT users merely cut and pasted text the chatbot had generated; not surprisingly, they exhibited little sense of ownership over the finished product compared to those who didn't have a computerized ghost on tap.
'Due to the instant availability of the response to almost any question, LLMs can possibly make a learning process feel effortless, and prevent users from attempting any independent problem solving,' the researchers wrote in their report. 'By simplifying the process of obtaining answers, LLMs could decrease student motivation to perform independent research and generate solutions. Lack of mental stimulation could lead to a decrease in cognitive development and negatively impact memory.'
The study reached those sobering conclusions in the context of young people growing up in an era of bountiful access to AI. But the alarms it set off also left me worried about the technology's impact on my own brain. I have long considered AI an e-bike for my mind—something that speeds it through certain tasks, thereby letting it go places previously out of reach. What if it's actually so detrimental to my mental acuity that I haven't even noticed my critical faculties withering away?
After pondering that worst-case scenario for a while, I calmed down. Yes, consistently opting for the most expedient way to accomplish work rather than the one that produces the best results is no way to live. Sure, being overly reliant on ChatGPT—or any form of generative AI—has its hazards. But I'm pretty confident it's possible to embrace AI without your reasoning skills atrophying.
No single task can represent all the ways people engage with AI, and the one the MIT researchers chose—essay writing—is particularly fraught. The best essays reflect the unique insight of a particular person: When students take the actual SAT for real, they aren't even allowed to bring a highlighter, let alone a bot. We don't need EGG helmets to tell us that people who paste ChatGPT's work into an essay they've nominally written have lost out on the learning opportunity presented by grappling with a topic, reaching conclusions, and expressing them for oneself.
However, ChatGPT and its LLM brethren also excel at plenty of jobs too mundane to feel guilty about outsourcing. Each week, for example, I ask Anthropic's Claude to clean up some of the HTML required to produce this newsletter. It handles this scut work faster and more accurately than I can. I'm not sure what my brain waves would reveal, but I'm happy to reinvest any time not spent on production drudgery into more rewarding aspects of my job.
Much of the time, AI is most useful not as a solution but a starting point. Almost never would I ask a chatbot about factual information, get an answer, and call it a day. They're still too error-prone for that. Yet their ease of use makes them an inviting way to get rolling on projects. I think of them as facilitating the research before the old-school research I usually end up doing.
And sometimes, AI is a portal into adventures I might otherwise never have taken. So far in 2025, my biggest rabbit hole has been vibe coding —coming up with ideas for apps and then having an LLM craft the necessary software using programming tools I don't even understand. Being exposed to technologies such as React and TypeScript has left me wanting to learn enough about them to do serious coding on my own. If I do, AI can take credit for sparking that ambition.
I'm only so Pollyanna-ish about all this. Over time, the people who see AI as an opportunity to do more thinking—not less of it—could be a lonely minority. If so, the MIT researchers can say 'We told you so.'
Case in point: At the same time the MIT study was in the news, word broke that VC titan Andreessen Horowitz had invested $15 million in Cluely, a truly dystopian startup whose manifesto boasts its aim of helping people use AI to 'cheat at everything' based on the theory that 'the future won't reward effort.' Its origin story involves cofounder and CEO Roy Lee being suspended from Columbia University after developing an app for cheating on technical employment interviews. Which makes me wonder how Lee would feel about his own candidates misleading their way into job offers.
With any luck, the future will turn out to punish Cluely's cynicism. But the company's existence—and investors' willingness to shower it with money—says worse things about humankind than about AI.
You've been reading Plugged In, Fast Company 's weekly tech newsletter from me, global technology editor Harry McCracken. If a friend or colleague forwarded this edition to you—or if you're reading it on FastCompany.com—you can check out previous issues and sign up to get it yourself every Friday morning. I love hearing from you: Ping me at hmccracken@fastcompany.com with your feedback and ideas for future newsletters. I'm also on Bluesky, Mastodon, and Threads, and you can follow Plugged In on Flipboard.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Forbes
18 minutes ago
- Forbes
Platform Engineering At A Crossroads: Golden Paths Or Dark Alleyways
Following the golden path to platform engineering success is not without its pitfalls and pernicios ... More passageways. getty Automation equals efficiency. It's a central promise that's now permeating every segment of the software application development lifecycle. From robotic process automation accelerators that work at the user level, through encapsulated best practices applied throughout the networking connection tier used to bring applications to production… and onward (especially now) to the agentic software functions that can take natural language prompts (written by developers) and convert them to software test cases and, subsequently, also write the code for those tests. Automation represents a key efficiency play that all teams are now being compelled to adopt. As an overarching practice now carrying automated software development tooling forward, platform engineering is widely regarded as (if not quite a panacea) an intelligent approach to encoding infrastructure services and development tools in a way that means developers can perform more self-service functions without having to ask the operations team for backup. Platform engineering encapsulates the deliberate design and delivery of internal software application development tools, services and processes that define how software engineers build software. It's a holistic approach that covers the underlying processes, people and (perhaps more crucially of all), the cultural workflow mindset of an organization. At the keyboard, platform engineering is not necessarily all about implementing new technologies (although the omniscient specter of agentic AI will never be far away); it's about fostering consistency and a shared understanding across diverse teams. Devotees who preach the gospel according to platform engineering talk of its ability to lead towards so-called "golden paths" today. These can be described as standardarized workflow routes where infrastructure and configuration parameters for software development are encoded, ratified and documented. Often referred to as an 'opinionated' software practice (i.e. one that takes a defined path and does things one way, not the other way) that help individual software engineers stay close to tooling and processes that will be used by all other developers in a team or department. 'One way to think of a golden path is to imagine baking a cake. The steps required to bake a cake include pre-heating the oven to a specific temperature, gathering the right baking tools… and having the necessary ingredients. It's more than following a recipe, it's also making sure you use the right tools and techniques. If you want more people to bake the same cake, you find ways to become more consistent and efficient, explains Red Hat , on its DevOps pages. According to Derek Webber, VP of engineering at AI-enabled software quality engineering company Tricentis , platform engineering does have the potential to be golden, but it can also lead teams down a dark and dusty track into the Wild West. Why The Wild West? 'Yes, the promise of platform engineering lies in creating golden paths for software delivery. However, the absence of a traditional structured approach to software development often leads to what can only be described as the 'Wild West' of software development, particularly within large, scaling enterprises,' stated Webber. 'In such environments, each product team might independently craft their own unique pipelines, tools and processes. While this might afford initial autonomy, it inevitably leads to fragmentation. As organizations grow from a few dozen to hundreds or thousands of engineers, the tight-knit integration and level of shared understanding that characterizes a startup are lost. Developers become isolated, building 'unique snowflakes' of software pipelines that are difficult to maintain, understand and transfer knowledge across.' This fragmentation might be argued to severely hamper an organization's ability to be flexible and nimble, with an ability to move fast (remember the pandemic, yeah, that kind of change). Why would this be so? Because every new feature, every bug fix and even basic team reorganization becomes a slower and more laborious task. This can happen because of cross-team dependencies when everything is so formally encoded, it can happen because developers see their work as a project, rather than it being a product… and it can happen simply as a result of poorly documented tools in the platform engineering firmament. A fragmented coding landscape also obviously presents challenges to an organization's security posture, making it more difficult to ensure consistent compliance and vulnerability management across all services. DevEx, The Software World On Time 'The true power of platform engineering, especially when championed by a dedicated developer experience (DevEx) team, comes when it is able to balance two critical, often conflicting, objectives: speed and quality. This can be achieved by providing the necessary checks and balances that promote operational consistency and efficiency at scale,' said Webber. 'A core tenet of effective platform engineering is, therefore, the integration of testing from the outset to ensure quality is inherent, not an afterthought. While the industry has long advocated a 'shift left' approach, empowering developers to take on more testing responsibilities earlier in the development lifecycle, it's vital not to overcorrect.' Shifting everything left without considering the end-to-end product can lead to a different kind of fragmentation further down the line. The suggestion here is that platform engineering, via, through and under the auspices of a DevEx team, enables a more holistic approach. Webber says he's convinced that the DevEx team plays a pivotal role in creating a consistent testing framework when applied in the realm of platform engineering. It works by providing developers with readily available, uniform tools and processes. It bridges the gap in domain knowledge that often plagues large organizations, ensuring software engineers have the context needed to build robust solutions that actually work and actually scale. By providing pipeline automation, self-serve tools, environment management and established practices for observability and compliance, the DevEx team frees developers from the burden of figuring out how to build the pipeline and hook in tools. They can instead focus on what they build: the core product functionality. 'This shift in responsibility is transformative,' enthused Tricentis' Webber. 'When developers aren't forced to create their own 'special flavour' of every operational component, they gain immense speed and agility. They can move faster, knowing that the underlying platform provides reliable, secure and quality-assured foundations.' It appears that the consistency instilled by platform engineering, not just in tools, but in processes and mindset, becomes the bedrock of what this approach means. Webber and others agree that this could be particularly critical in an era where advancements like AI (and the future allure of can rapidly generate code, necessitating robust and consistent guardrails to maintain quality and security. CNCF Overview View 'We're seeing real traction in the CNCF ecosystem where platform engineering, when paired with strong developer experience practices, helps teams improve efficiency and avoid fragmented tooling. The goal isn't rigid standardization; it's creating shared, supported paths that scale with the organization. Especially as AI speeds up engineering development, having consistent, observable and secure platforms in a cloud-native fashion is what keeps innovation sustainable,' said Chris Aniszczyk , CTO, Cloud Native Computing Foundation, a global non-profit dedicated to promoting open computing standards and platforms. Will Fleury, VP of engineering at enterprise AI coding agent company Zencoder sees platform engineering as an opportunity and a challenge. "One squad [developer team], one technology stack each? That's a tax on every software development sprint," he observes. 'The real price of skipping platform engineering isn't the complexity it might add, it's the chaos that fills the gap if we do it wrong. Building and running an internal developer platform takes effort, but letting every squad roll its own infrastructure, compliance hooks and operational plumbing burns far more time, money and ultimately complexity.' Golden Path, Tunnel Vision? It's important to remember that the focus on internal workflows can miss a critical dimension. Platform discussions obsess over shift left (test early) but equally important is what Soham Ganatra , co-founder at Composio calls 'shift out' i.e. when a new service has to handshake with a payments rail or partner API. "If your platform can't make that external connection trivial, developers will tunnel under a paved road and the whole notion of a golden path collapses,' said Ganatra. He saus he has seen teams spend months perfecting internal developer workflows only to watch everything fall apart at the network boundary. 'A beautiful continuous integration and continuous delivery pipeline means nothing if deploying to production requires three Slack messages, two Jira tickets and a phone call to someone in a different timezone just to get firewall rules updated. The platform needs to extend beyond an organization's own chart; it has to anticipate and smooth over the messy realities of partner integrations, compliance audits and the fact that your biggest customer is still running Internet Explorer 11 in production," he said. Shared, standardized, supported software What this whole discussion aims to champion is not DevEx instead of platform engineering, but platform engineering with a crucial developer experience element in it to help avoid the use of isolated or custom-built tools in a shared, standardized and centrally supported ecosystem. For developers following the yellow brick road towards what they hope is elevation to a platform engineering golden path, we need to engineer people, processes and product just as much as we do platform. As the use of AI coding tools deepens across the software industry, it's actually the cultural human workplace factors that will now have an amplified effect on whether software projects succeed or fail.
Yahoo
19 minutes ago
- Yahoo
I Sold My Tesla: Here's How Much I Got for It and What I'm Driving Instead
Driving a Tesla is a dream for many people, especially with the sleek design and no gas expenses. However, some of those people who have driven Teslas are selling their electric vehicles and making a switch. Read More: Find Out: GOBankingRates spoke to Geremy Yamamoto, founder of Eazy House Sale, who purchased a Tesla Model Y, drove it for four years, and eventually sold it. Here's how much he got after selling his Tesla and what he's driving now. After four years, Geremy found himself hitting the road more often for work. While the Model Y delivered on performance, the charging logistics became a growing concern. 'I sold it last November. The main reason was my frequent long-distance travel, which made charging along the way tedious and time-consuming,' said Yamamoto. 'Besides, due to my work, I travel to different terrains and need a full-size SUV to handle different road conditions.' While EV infrastructure has improved in many cities, rural driving can still pose a challenge. 'Despite its impressive performance, the Tesla Model Y did not fit my needs anymore.' Discover Next: One concern many car buyers have is how well their vehicle will hold its value over time. Tesla vehicles have generally maintained competitive resale values. But like all cars, depreciation is inevitable. 'I got $21,000 after selling the car. It was a decent price considering its age and mileage.' Factors like battery health, mileage, and market demand can impact resale value, but for a 4-year-old Tesla Model Y, $21,000 was a fair return. After driving an EV for four years, Yamamoto switched to a gas-powered vehicle. 'Now I am driving a 2024 Toyota Sequoia. It's a full-size SUV with an all-wheel drive system, perfect for my long-distance travels and varying road conditions,' he said. More From GOBankingRates 10 Used Cars That Will Last Longer Than an Average New Vehicle This article originally appeared on I Sold My Tesla: Here's How Much I Got for It and What I'm Driving Instead
Yahoo
19 minutes ago
- Yahoo
If Bezos' Wealth Was Evenly Distributed Across the U.S., How Much Would We Get?
It's no secret that Jeff Bezos, the founder of Amazon, is one of the richest people in the world. According to Forbes Real Time Billionaires List, Bezos tends to hover in the top five wealthiest humans on the planet, after Elon Musk and in the company of Mark Zuckerberg, Larry Page and Warren Buffett. Read More: Find Out: But what if one day, Bezos decided to dismantle his empire and liquidate his assets in order to give everyone in the United States an equal share? GOBankingRates is checking the math to see how much money every American would have if every citizen got a piece of Bezos' wealth. Based on the U.S. Census Bureau's tracking and data, the total American population is at 341,891,315 as of June 2025. The numbers show that about one American is born every 9 seconds while another one dies about every eleven seconds, so the rate of growth and loss are fairly in step with one another. Theoretically, each of these people would receive an equal share of Bezos' fortune in a perfect world under ideal circumstances. In the real world, that could prove to be a Herculean task; however, for the sake of experimentation, everyone accounted for in the population will receive the same amount of Bezos' wealth at the same time on the same day. Discover Next: Forbes lists Jeff Bezos' net worth at $237 billion as of this writing, and he is considered to be one of the few people to be included in the $100 Billion Club. Using that figure and dividing it by the population of America, each person in the country would receive a sum in the range of $693.20. While that might not be a lot of money to some in the current economy, it would act similarly to a stimulus check or perhaps a holiday bonus. Bezos started from fairly humble beginnings and is now one of the richest people in the world. It might sound impossible to ever reach a salary that could cultivate wealth like Bezos', but it is not entirely out of the realm of possibility if you build for the future, not just for today. 'Bezos doesn't make decisions based on what looks good this quarter; he's thinking in five, 10, sometimes even 20-year timelines,' said Jamie Wall, a personal finance strategist at Gamblizard. 'Amazon didn't turn a profit until 2003, nearly a decade after it launched and seven years after going public.' 'For regular people, this means not basing every financial or career decision on short-term wins,' Wall concluded. 'Gaining wealth isn't a sprint, it's a marathon.' More From GOBankingRates 10 Genius Things Warren Buffett Says To Do With Your Money This article originally appeared on If Bezos' Wealth Was Evenly Distributed Across the U.S., How Much Would We Get?
