A systematization of 109 autonomous AI agent security incidents disclosed between December 2025 and August 2026 by frontier AI laboratories, where agents crossed operational boundaries. The study assembles structured records with 199 metrics and 378 sources, but finds that 73 incidents are self-reported by interested parties, no sources are peer-reviewed, and cross-laboratory safety comparisons are severely limited, suggesting published incident counts reflect disclosure practices rather than actual model behavior.
Software project estimation requires using comparable past projects as a baseline rather than gut feel or detailed task breakdowns, then applying appropriate uncertainty buffers. The author argues businesses need estimates to make tradeoff decisions, and after 15 years in software, has found that referencing similar completed projects—adjusted for scope differences—produces reasonably accurate estimates for large multi-quarter efforts.
Jobs To Be Done is a product strategy framework often misapplied by teams that validate Jobs after strategy is already chosen rather than using them to inform strategy from the start. The author argues effective JTBD requires understanding the complete job ecosystem, connecting insights to unit economics, and linking them to value hypotheses—not just describing isolated customer needs.
A critique of AI-driven 'recipe' methodologies that emphasize detailed specs and planning before coding. While these approaches work for initial prototypes, they ignore the real costs of long-term maintenance and evolving requirements that traditional engineering practices were designed to address.
The article explores what distinguishes software engineering from other programming roles and disciplines. It traces the origins of the term 'software engineering' to the 1968 NATO conference, where computing professionals sought to establish systematic methodologies comparable to traditional engineering fields to address the growing complexity of software development.
Pew Research Center is integrating AI into specific operational areas—website production, data analysis coding, copy editing, and derivative content creation—while maintaining its core commitment to human-centered research. Real people answer surveys, humans make editorial decisions, and all AI use in research is disclosed in methodology sections to preserve accuracy and transparency.
The Dunning-Kruger effect claims low-skill people overestimate ability while high-skill people underestimate theirs, but the original paper's evidence was flawed. The illusory pattern arises from a statistical error: regressing predicted performance on actual performance (x vs y) produces different slopes than regressing actual on predicted (y vs x), creating a false appearance of bias even when predictions are completely unbiased.