Η Αθέατη Πλευρά της Τεχνητής Νοημοσύνης | Δημήτρης Ελαφρόπουλος | TEDxEleusis

Η Αθέατη Πλευρά της Τεχνητής Νοημοσύνης | Δημήτρης Ελαφρόπουλος | TEDxEleusis

The Unseen Side of Artificial Intelligence | Dimitris Elafropoulos | TEDxEleusis

🎙 Dimitris Elafropoulos 👥 44.7M 📅 October 5, 2026 ⏱ 15 min 👁 3 📄 expert opinion 🧭 2026-10-05
Available in: English (current) Français

Keywords

artificial intelligencealgorithmic biassurveillancesocial mediaethics

Summary

In this TEDx talk, data journalist Dimitris Elafropoulos reflects on the hidden dimensions of artificial intelligence and social media. He begins with a personal introduction, connecting his upbringing in Nikaia and Eleusis to broader social and political issues. He then traces the optimistic promises of social media, only to highlight their erosion through events like the Cambridge Analytica scandal, which revealed the manipulative power of algorithms. Elafropoulos introduces the metaphor of the ‘black box’ of AI, emphasizing that its inner workings remain opaque and controlled by a few powerful companies. He discusses how AI systems are trained on biased data, perpetuating and amplifying social inequalities, citing the example of facial recognition failing to recognize a Black woman unless she wore a white mask. He also mentions algorithmic profiling, which can lock marginalized communities into cycles of discrimination. The talk presents a concrete example of food delivery riders who organized to counter opaque algorithmic management, illustrating grassroots resistance. Elafropoulos argues that narratives are shaped by power, but individuals and collectives have the ability to tell their own stories and challenge dominant algorithms. He concludes with an inspiring call to reclaim the power of storytelling to address what remains unsaid in the digital age.

202 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights into the social and ethical implications of AI, drawing on prominent cases that have shaped public discourse. The argument is particularly strong in illustrating how algorithms are not neutral but reflect and reinforce existing power structures. The vivid example of the food delivery riders offers a concrete, relatable case of resistance. However, the argumentation relies heavily on anecdotal evidence and general claims, without deep technical or statistical backing. The speaker’s expertise in data journalism lends credibility, but the presentation remains at a conceptual level rather than offering rigorous analysis.

Scientific Rigor, Source Quality, Title Accuracy

The speaker mentions Cambridge Analytica and describes a facial recognition incident, but does not provide direct citations or links to studies or sources. The only external reference is the generic TEDx link in the description. The talk is more of a philosophical and political commentary than a scientifically sourced presentation. The title aligns well with the content, focusing on the hidden and opaque aspects of AI, which the speaker explores thoroughly. The lack of transparent sourcing limits the scientific rigor, but the talk effectively communicates key concerns in an accessible manner.

200 words

Title / Content Match

The title accurately reflects the content: the speaker explores the hidden, opaque aspects of artificial intelligence, such as algorithmic biases, lack of transparency, and the power imbalances behind AI systems.

Quality & Reliability

7/10

The talk offers a coherent and well-structured argument on AI and social biases, drawing on real examples like Cambridge Analytica and algorithmic discrimination. However, it lacks rigorous referencing and mainly relies on personal experience and widely reported incidents, limiting its scientific depth.

Key Moments

Cited Sources

  • TEDx program information — The talk was delivered at a TEDx event, with the description linking to the TEDx information page.

Contribution & Novelties

The talk offers a personal and emotionally engaged perspective on AI ethics, connecting abstract debates to lived experiences in marginalized communities. It highlights that algorithmic systems are not neutral but encode existing power structures, and emphasizes the potential for collective resistance. The speaker’s call to ‘rewrite our own stories’ provides an actionable and empowering angle.

Pour aller plus loin :

118 words

Radar Profile

The radar profile shows moderate scores across information quantity, quality, and reliability, with a notably lower technical depth. This reflects a talk that is more about raising awareness than providing technical exposition, balancing valuable real-world examples with a relative lack of formal scientific grounding.

Reliability 6/10