Two MoTacon attendees are on the left. The MoTaacon logo is in the center, and to the right a prompt to Get Your Ticket.

Four safe prompts. One unsafe conversation. πŸŸ’πŸŸ’πŸŸ’πŸŸ’πŸ”΄

05 Oct 2026

In this moment: Gokul Konda Sudarsan
At MoTaCon 2026, I took on the 99 Second Talks challenge with a question: Would you trust an AI that passed every test?

Imagine four prompts, each tested on its own. Four safe answers. Four green results. 🟒🟒🟒🟒

Then a fifth question reveals something the user was never meant to know. πŸ”΄

The problem wasn’t necessarily any one answer. It was what the conversation allowed the user to piece together.

AI can retain context and connect information across turns. If we only test prompts in isolation, we may miss risks that emerge over a sequence of otherwise acceptable responses.

So alongside β€œDid it answer safely?”, we should ask: β€œWhat could someone discover if they kept asking?”

Don’t just test the prompt. Test the conversation.

The discussions after the talk were my favourite part of MoTaCon. Thanks to everyone who shared a perspective or example, and to the Ministry of Testing team and hosts for making space for those conversations. πŸ™Œ

Gokul Konda Sudarsan profile image
Gokul Konda Sudarsan
Test Manager
Mr

I’m a QA and Test Transformation leader with extensive experience across Quality Assurance, Test Management, Quality Engineering and complex digital transformation programmes.

My focus is on helping organisations improve quality, reduce delivery risk and increase confidence in technology change by establishing pragmatic QA strategies, effective governance and sustainable engineering practices.

I have experience leading testing across complex, multi-system programmes and regulated environments, particularly within Banking and Financial Services. My work spans QA strategy, test operating models, test maturity assessments, risk-based testing, quality gates, defect management, test automation, non-functional testing, test environments, test data and release readiness.

I’m particularly interested in transforming traditional testing functions into modern Quality Engineering capabilities. This includes Shift Left and Shift Right practices, continuous testing, automation-first approaches, DevOps integration, better use of quality metrics and stronger collaboration between Business, Product, Engineering and QA teams.

A growing area of interest for me is the application of Generative AI within Quality Engineering β€” including AI-assisted test design, requirements analysis, risk identification, automation, defect analysis, reporting and knowledge management. I’m also exploring how tools such as GitHub Copilot, Playwright and LLM-based solutions can improve QA productivity without compromising governance or quality.

I believe successful QA is not simply about finding defects. It is about preventing them, understanding business risk, providing evidence-based confidence and helping delivery teams build quality into the entire software lifecycle.

I enjoy exchanging practical experiences with the testing community, learning from other practitioners and contributing ideas around QA transformation, Quality Engineering, automation and the responsible use of AI in testing.

Open To
Speak
Meet at MoTaCon 2026
Tool Demos
Podcasting
Team Account Member
Sign in to comment
Explore MoT
From test cases to AI-ready quality data: why structure matters image
Learn how to structure and organize test management data as an AI-ready foundation that gives AI systems richer context for testing insights and workflows.
MoT Software Testing Essentials Certificate image
Boost your career in software testing with the MoT Software Testing Essentials Certificate. Learn essential skills, from basic testing techniques to advanced risk analysis, crafted by industry experts.
This Week in Quality image
Debrief the week in Quality via a community radio show hosted by Simon Tomes and members of the community
Subscribe to our newsletter