"The developers I talked to agreed that LLMs will stick around and play a role in programming in the future in some fashion, but worried about how the industry will adapt to executives’ current obsession with the technology, especially when it comes to fostering future generations of developers. “Older programmers will be fine if there are any jobs left in a few years, but I worry for people early in their careers,” the UX designer told me. “We are hiring junior programmers who rely on AI to complete the simplest tasks. They don't have the knowledge or experience to know when AI output is error-laden or inefficient.”“I wish I had a crystal ball for this one, but my gut feeling is that this method of building software will be unsustainable either economically or in terms of tech debt,” the software engineer at the FAANG company said. “There's a pretty clear split on my team between people who love AI coding and those who just do it because it's what the company wants, and generally speaking I find that the people who are still [technically focused individual contributors] with their nose in code all the time are less likely to be big AI boosters. I think the tech and its outputs start to really break down the more you question them and those who are doing that day in and day out tend to have a worse opinion of the tech.”“I think there will be a ‘reckoning’ or ‘awakening’ from the industry notion that now everyone can code and that vibe coding is viable for a real production app and software companies are dead,” the developer in fintech said. “I think we will grow to find the patterns and industry best practices that will balance the negatives of LLM development (hallucination, unstructured code) with better techniques to verify the output's correctness at scale, and the hype and techno optimism of AI will get to a saner middle ground.”"404media.co/software-developer#AI #GenerativeAI #SoftwareDevelopment #Programming #AICoding #LLMs
On the other hand...: "In February 2026, we estimated that frontier models’ 80%-reliability cyber time horizon had doubled every 4.7 months since reasoning models emerged in late 2024, given a 2.5M token limit. This was around half our November 2025 doubling time estimate, which was 8 months for both 50% and 80% reliability. Claude Mythos Preview and GPT-5.5 have since significantly outperformed this trend. At the time of writing, it's unclear whether Mythos Preview and GPT-5.5 represent an isolated break from existing rates of progress or are part of a new, faster trend.Mythos Preview and GPT-5.5 have large upper-bound error bars due to near-100% success rates on our narrow cyber suite’s longest tasks, even with the 2.5M token limit.2 Our tasks are also not long enough to determine how sharply the models’ reliability would deteriorate at higher task lengths. This places some of the latest models at the limit of what our narrow test suite can measure. Without the 2.5M token cap, success rates are so high that time horizons become impossible to calculate. The cap, alongside our use of a simple agent scaffold, artificially lowers success rates and understates what models can do with more tokens and stronger scaffolds. In return, it ensures time horizons are measurable and can be compared across models. For reference, a 2.5M token limit is relatively low – in our cyber range experiment we use up to 100M tokens and find performance would likely still improve beyond that budget, especially for recent models, which disproportionately benefit from higher token limits."aisi.gov.uk/blog/how-fast-is-a#AI #CyberSecurity #GenerativeAI #LLMs