Anthropic: los expertos siguen a la IA el 86 % de las veces sin darse cuenta

Value of the Information & Strength of the Argument

The video provides a valuable perspective on a real and relevant study, highlighting the subtle ways AI can influence decision-making. The argumentation is personal and reflective, connecting the study’s findings to everyday AI use. It offers a practical solution (the ‘judgment call’ method) to mitigate automation bias. However, the presenter does not critically engage with the study’s limitations beyond noting the small sample size, and his argument would benefit from more rigorous statistical and methodological questioning. The discussion is engaging but relies heavily on anecdotal experience.

Scientific Rigor, Source Quality, Title Accuracy

The video cites the primary source (Anthropic research page) and links to the full episode. The title accurately represents the video’s focus on the 86% result and the lack of awareness. The presenter’s interpretation is subjective but transparently so. No external contradicting sources are mentioned, and the video does not provide a thorough scientific breakdown, though it remains honest about uncertainties. The overall rigor is moderate, consistent with an expert opinion piece.

The title accurately reflects the main finding discussed (86% adherence) and the unnoticed influence, matching the content.

7/10

The video references a specific study by Anthropic and provides a link, but the analysis is subjective and includes personal opinions. It lacks rigorous statistical critique and generalizes from a small sample.

Markers derived by PSI from the transcript: the creator did not define chapters.

• 00:00Introduction and clickbait headline about the 86% finding.
• 00:30Explanation of the experiment with 61 analysts and the coin toss methodology.
• 02:00Discussion of the 86% adherence and 15% recognition rates, and initial reaction.
• 04:00Analysis of why a lack of correct answer allows AI persuasion.
• 05:00Introducing the 'judgment call' method to counter bias.
• 06:40Personal experience with using Obsidian and separating human thought from AI-generated text.
• 08:00Example of analyzing a video claim by downloading the transcript and fact-checking.
• 09:40Conclusion emphasizing the need for intentional critical thinking when using AI.

• Anthropic Research: Agents and Judgment Calls — Primary source of the study discussed, describing the experiment and findings.
• Full episode of El Test de Turing (from this point) — Link to the full episode for additional context.

• Anthropic Research: Agents and Judgment Calls — Directly supports the video's central claim about expert adherence to AI suggestions.