Editorial policy
Last updated 29 October 2026.
What we cover
Security and AI news: vulnerabilities under attack, data breaches and incidents, attacks on AI systems and agents, AI model launches and the claims made for them, and the regulation and guidance that follows.
How we write the news
pk-sharma.com reports security and AI news from the original sources: vendor advisories, government notices, researchers’ reports and company statements. We read the source, write the story in our own words, and link the source so you can read it yourself. AI helps us do this.
What every briefing follows
- Every figure, date, version number and quotation comes from a source we read, and that source is linked in the article.
- A figure we calculated rather than found is labelled “derived”.
- Claims by companies, researchers and attackers are credited to them. An allegation stays an allegation until someone with the evidence confirms it.
- Each briefing says what the sources do not establish, as well as what they do.
- Where a briefing relies on a company’s own claims, it says what that company sells.
- Private individuals are not named. People appear by their role.
Images
Main images are made with an AI image model. Diagrams are drawn from the sources. Screenshots and figures from vendors or researchers are credited to them.
Corrections
When something is wrong, we fix it and list the change, with its date, on the corrections log. How that works is set out in the corrections policy. To report an error, email contact@pk-sharma.com.
Who publishes this site
pk-sharma.com is published by Parminder Kumar Sharma (P.K. Sharma) from Cambridge, United Kingdom. More about him is on the about page, and the ways to get in touch are on the contact page.