AI's 'Development Revolution' Unmasked: Devs Point to Dijkstra and OWASP to Call Out Corporate Hype

Post date: April 7, 2026 · Discovered: April 18, 2026 · 5 posts, 41 comments

The current push to solve complex software engineering challenges using LLMs like Claude faces immediate skepticism. The consensus rejects the idea that natural language prompting can replace rigorous coding practices, citing the known difficulty of security compliance like OWASP ASVS.

Commenters argue AI adoption is more of a workforce reduction strategy than an efficiency gain. 'chocrates' stated management seeks 'getting rid of technical folks entirely,' not fixing processes. 'jtrek' added that prompt writing and output checking simply create 'process slog.' Regarding security, 'underisk' noted executives lack deep technical knowledge, making even referencing specific standards imprecise.

The raw takes show deep division: one segment dismissed the source code leak as irrelevant, pointing out the code was already public on npm, while others treated it like a major vulnerability event. The prevailing view is that the hype vastly overstates AI's capability, suggesting complex engineering tasks remain stubbornly resistant to simple English prompts.

Key Points

OPPOSE

LLMs fail to address core engineering complexity.

The difficulty of secure coding practices, like meeting OWASP ASVS, proves LLMs are insufficient replacements for actual programming (addie).

OPPOSE

AI is a threat to technical jobs, not just a tool.

chocrates argues that company interest is 'removing technical labor entirely' rather than assisting skilled staff.

OPPOSE

Over-reliance on prompts introduces new work slowdowns.

jtrek contends that prompt writing and subsequent output checking create more bureaucratic 'process slog' than they save time.

OPPOSE

Natural language programming is conceptually flawed.

Multiple users referenced deep theory, like Dijkstra's 1978 critique, against English-based logic.

MIXED

The reported 'code leak' was largely noise.

Some treated the leak as a major breach, while others correctly pointed out the code was already accessible on npm (artwork on [email protected]).

Source Discussions (5)

This report was synthesized from the following Lemmy discussions, ranked by community score.

452
points
Hey we solved software development, no need to learn programming anymore (Claude's source code leak)
[email protected]·41 comments·4/7/2026·by tired_n_bored·lemmy.world
39
points
Claude Code source leak reveals how much info Anthropic can hoover up about you and your system
[email protected]·0 comments·4/1/2026·by StopTech·theregister.com
18
points
AfterPack: Claude Code's Source Didn't Leak. It Was Already Public for Years.
[email protected]·1 comments·4/2/2026·by artwork·lemmy.world
16
points
They thought they were downloading Claude Code source. They got a nasty dose of malware instead
[email protected]·0 comments·4/2/2026·by lemmydev2·theregister.com
11
points
Claude Code Found a Linux Vulnerability Hidden for 23 Years
[email protected]·0 comments·4/3/2026·by lemmydev2·mtlynch.io