Beyond the Beach: News That Shaped Summer 2025

I really tried my best to unplug during my summer break, but even without a computer, my phone kept me tethered to the news.

One particular story caught my attention and stuck with me: an MIT study on the impact of AI on businesses.

It really got me thinking about the wider implications for the tech community

The Reality of the "AI Gold Rush”

As reported by Fortune, the MIT study revealed a shocking statistic: 95% of AI pilot programs fail to increase revenue.

This was a huge surprise, especially given the sheer amount of money recently poured into AI development. The current frenzy feels like a true "AI Gold Rush," reminiscent of the 1800s, where everyone is rushing to strike it rich without a clear strategy.

Why AI Pilots Fail

The MIT study explained that these failures happen because businesses are simply inserting new AI tools into existing workflows without making fundamental changes.

This classic mistake is similar to what happened in the 1990s when companies simply swapped typewriters for personal computers and expected an instant digital transformation. To truly benefit from innovation and see a return on investment, companies need to revolutionize their operations from the inside out. The same is true for AI.

To make this transformation successful, new experts are needed, like prompt engineers and validation engineers. A validation engineer is a new role responsible for checking the output of an AI tool, flagging errors or "hallucinations," and providing feedback to the prompt engineers. Think of them as the essential "middleman" who ensures accuracy and quality control.

AI’s Impact on the Workplace

While an AI tool can help an employee write an email, the minimal time saved rarely justifies the high cost of implementation. The real return on investment comes from achieving much more significant financial gains.

Another example is the use of AI for coding. Letting every engineer freely use AI tools to generate corporate code would lead to chaos. Corporations have strict standards for compliance, security, and quality, including things like code style, documentation, and automated testing. For AI to work effectively in this environment, companies need to provide prompt templates that ensure the generated code aligns with these standards.

On top of this, an engineer must now read, debug, and modify code written by an AI. This is a massive challenge that no one enjoys, especially senior engineers. Think about the code running the Voyager satellite, which is still maintained decades after it was written! This kind of code maintenance has always been a nightmare, and AI doesn’t change that.

While AI tools will undoubtedly provide long-term benefits for coding, this will only happen if companies update their entire Software Development Life Cycle (SDLC). The skills needed for coding engineers in this new AI era are very different and require a specific seniority. We're already seeing layoffs of junior software engineers at companies like Salesforce, which have adopted AI-based coding tools.

It seems the market's expectation for immediate revenue increases from AI was a bit too optimistic.

The Path Forward

It's easy to get swept up in this new AI gold rush, but barreling forward without a clear plan will likely lead to disaster, resulting in high costs and minimal benefits, as the MIT study shows.

The solution is to slow down, do your homework, and carefully analyze how AI can be integrated into your company's workflow.

By taking the time to do the proper diligence, companies would be better prepared for the new challenges the market will throw at them and see the real benefits these tools can provide.

Will the companies in question actually do this? Who knows. It definitely doesn’t seem likely right now, when AI is the shiny new thing all the “cool kids” are playing with and tech behemoths are all falling over each other to see who gets to the mountaintop first.

When they start really seeing the effects to their bottom line though? I suspect a significant shift in attitude is on the horizon.