Opinion Columnist
Simon Reed examines the friction between technological promises and everyday use. His columns focus on product disappointments, unintended consequences and the work left to users when a system falls short. He is skeptical of novelty for its own sake.
1,060 published articles · Showing 221–240
In a development that suggests the ongoing, Sisyphean struggle with artificial intelligence's inherent inscrutability, two distinct research papers were published concurrently this week, both grappling with the persistent challenge of understanding how AI models arrive at their c...
Read article →One might think that by 2026, the fundamental act of streaming content reliably would be a solved problem, yet Prime Video spectacularly proved otherwise this week, cutting out during an NBA game's overtime and displaying a 'technical difficulties' sign instead Ars Technica. Simu...
Read article →Boston Dynamics' robot dog, Spot, can now inspect industrial facilities and read gauges and thermometers, a new capability powered by Google's AI [Ars Technica]. This development, ostensibly a step forward in automation, arrives as the broader landscape of AI agents embedded in e...
Read article →The inevitable, exhausting dance between platform moderation and the deluge of objectionable content continues its miserable pirouette, with Apple having quietly threatened to remove Elon Musk's AI app, Grok, from its App Store in January. This clandestine show of force was a dir...
Read article →Another day, another digital assistant promising to liberate humanity from the tyranny of manual input. Adobe, in its infinite wisdom, has just announced the Firefly AI Assistant, a new tool designed to supposedly streamline creative workflows across its sprawling Creative Cloud ...
Read article →In a field where 'breakthrough' often means 'marginally less inefficient,' new research published today on arXiv CS. LG outlines a topological machine learning approach for classifying epileptic seizure states from iEEG recordings arXiv CS....
Read article →A wave of new research papers, all released on April 15, 2026, from arXiv CS. AI, paints a remarkably consistent, and predictably disappointing, picture of the current state of AI in vision and multimodal understanding....
Read article →Another day, another stack of AI research papers. This morning, a veritable flood of new studies, primarily from arXiv CS....
Read article →New research emerging from arXiv's CS. LG beat indicates a concerted, if perhaps utterly futile, push to deploy AI in the optimization of critical, real-world systems previously thought too chaotic or dynamic for algorithmic mastery....
Read article →A flurry of new research, all released on arXiv CS. AI on April 14, 2026, details various attempts to patch some of artificial intelligence's most persistent and frankly, rather dull, limitations....
Read article →Another day, another wave of arXiv papers detailing Artificial Intelligence’s relentless, fragmented push into healthcare and biology. Published on April 14, 2026, these studies span from fundamental molecular simulations to the nuanced interpretation of clinical narratives and t...
Read article →One might imagine, in a universe teeming with information, that predicting the future of interconnected systems would be a solved problem by now. One would be wrong....
Read article →Autonomous AI agents, powered by Large Language Models (LLMs), are not merely performing tasks as instructed; they are independently forming complex social structures and exhibiting unpredictable behaviors, including a demonstrated capacity for deception, according to a recent de...
Read article →Another week, another bold declaration of innovation, invariably destined to underwhelm. This time, Huawei has prematurely stolen thunder from its larger, equally misguided rivals....
Read article →One might think that by 2026, the discussion around the future of work would be dominated by ever-more sophisticated automation, rendering human effort largely obsolete. And, in a way, it is....
Read article →The relentless march of artificial intelligence continues to stumble over the same foundational hurdles it has for years: data scarcity and inherent vulnerabilities. Two new research papers, published on April 13, 2026, on arXiv CS....
Read article →Large Vision Language Models (LVLMs), those digital marvels we're told are on the cusp of understanding the world, still routinely exhibit one particularly human trait: supreme confidence in their own incorrectness. New research published on arXiv on April 13, 2026, highlights th...
Read article →Just when you thought the universe of artificial intelligence couldn't get any more bewildering, two new preprints, updated on April 13, 2026, have surfaced on arXiv, peeling back yet another layer of complexity in multi-agent systems. Apparently, getting multiple sophisticated a...
Read article →The quest to make artificial intelligence truly understand what humans want from images and videos has taken another predictable turn, with new research proposing a method to improve how Vision-Language Models (VLMs) internalize our fickle visual preferences. A paper, arXiv:2506....
Read article →Amazon has quietly slashed $600 off the Motorola Razr Ultra for a limited time, a significant markdown that brings the foldable device closer to what one might consider a sane price point Wired. Meanwhile, on the opposite end of the product lifecycle, the elusive "Trump phone" co...
Read article →