We Were Never Asked to Write Code. We Were Asked to Solve Problems.
EXANTE CTO Richard Forss draws on the dot-com crash and AI breakthroughs to explain why AI will help engineers solve bigger problems, not replace them.
In 2000, the author, Richard Forss, was working at Deutsche Bank on a new trading platform while watching friends leave for startups. The author felt envious of their success, but the financial crash soon showed that the promised riches were not materializing. The author credits the current mood to a recurring pattern in the industry, with "Artificial intelligence" being a marketing term coined in the 1950s by scientists seeking military funding.
The hype around AI leads to inflated expectations and eventual disappointment, as funding inevitably leaves when reality sets in.
The author draws a parallel between the AI hype cycle and their own experience at Deutsche Bank. While deep learning made significant strides at the ImageNet competition in 2012, the author emphasizes that the gap between real capability and honest expectation remains. They caution against overselling and the inevitable "AI winters" that follow the hype.
In the healthcare sector, the author highlights Geoffrey Hinton's mistaken prediction that radiologists would be obsolete within five years due to deep learning. The author argues that instead of replacing human professionals, the tools have accelerated demand for their services. Despite the availability of agentic coding assistants in 2025, the author points out that engineers remain essential, as the technology still requires human oversight to differentiate between plausible and correct answers, particularly in critical situations.
The author concludes by emphasizing that the purpose of hiring engineers has always been to solve problems, not just to write code. The story serves as a reminder that the real value lies in addressing real-world challenges, regardless of the technology used.
Written by urgent.news from HackerNoon's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
