Author
Olga Fedotova
Head of Design, INITE
Writes about the surface where a person — or an agent acting for one — meets an automated system: what a browser agent can actually operate, why a chatbot fails on its design rather than its model, and where the human has to stay in the loop for the system to be allowed to run at all.
4 articles
Published here
Why Your Last Chatbot Failed, and It Was Not the Model
Most bad chatbots were measured on deflection rate, which pays them to avoid handing over. The metric caused the experience your customers hated.
Aug 22, 2026 · 5 min
What Your AI Is Not Allowed to Decide
Safe AI in a real deployment is four operating rules about where the human sits. Here they are, with what each one costs and where the line gets drawn.
Aug 3, 2026 · 5 min
Browser-Agent-Ready SaaS: Making Your App Usable by Operator and Claude
Operator, ChatGPT Agent, Claude Computer Use can log in and check out on your site - if you let them. 2026 audit for selectors, forms, ARIA, captchas.
May 11, 2026 · 6 min
Responsible AI in 2026: The Compliance Reality Behind the Ethics Talk
AI ethics in 2026 is mostly compliance. EU AI Act, NIST AI RMF, ISO 42001 are now enforceable - the gap between principles and audit evidence is the work.
Oct 2, 2025 · 7 min
15-minute self-check
Fifteen minutes, no email required. It maps where your work actually goes and ranks what is worth automating first.
Start the 15-minute self-checkStarts immediately in the browser.
- Fee
- Free
- Length
- 15 minutes
You keep the ranked list of candidates either way.
Cookies
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.