AI Disclosure

How AI tools are used in producing content on pranoti.thesciencetalk.com, and how that maps to Article 50 of the EU AI Act.

I write about AI integration, so it would be strange not to say plainly how I use it here.

How AI is used on this site

Articles in Perspectives, AI Guides and Case Studies are drafted with the assistance of AI tools, principally Claude. Every piece is reviewed, edited and fact-checked at drafting stage, and only then is it published. The judgement, the argument, the recommendations and the errors are mine.

Specifically:

  • Editorial responsibility. I, Dr. Pranoti Kshirsagar, hold editorial responsibility for all content published on this site. The Science Talk is the trading name of my sole proprietorship, not a separate legal person. Contact details are on the Imprint.
  • Claims and sources. Factual claims — particularly about regulation, funding programmes and tool behaviour — are verified against primary sources, and those sources are linked so you can check them yourself.
  • Workflows and results. Case studies describe automations I have actually built and run. Numbers reported in them come from real runs, not estimates.
  • Images. Where an image is AI-generated, it is labelled as such in the caption or alt text.
  • Your data. No personal data of site visitors is entered into AI systems. See the Privacy Policy.

Why this page exists

Article 50(4) of the EU AI Act (Regulation (EU) 2024/1689), which has applied since 2 August 2026, requires deployers of AI systems that generate or manipulate text published to inform the public on matters of public interest to disclose that the text was artificially generated or manipulated.

That obligation does not apply where the AI-generated content has undergone a process of human review or editorial control and a natural or legal person holds editorial responsibility for the publication. That is the case for every article here.

This page is therefore published voluntarily rather than because a disclosure is mandated. Readers deciding what weight to give an argument should know how it was produced.

Code of Practice on Transparency of AI-Generated Content

I have signed, as The Science Talk, the European Commission’s Code of Practice on Transparency of AI-generated Content, under Section 2, as a deployer of generative AI systems — signed ahead of the Article 50 obligations becoming legally binding on 2 August 2026. Signing commits me to marking AI-generated content appropriately and disclosing its use where it matters, consistently rather than only when required. Read more in The Science Talk Signs the EU Code of Practice on Transparency of AI-Generated Content.

What is not used

  • No chatbot or conversational AI system runs on this site. Were one added, it would identify itself as an AI system at the start of the conversation, as Article 50(1) requires.
  • No emotion recognition or biometric categorisation system is used (Article 50(3)).
  • No deepfakes or manipulated depictions of real people are published (Article 50(4)).

Sources

Last updated: 10 September 2026