Responsible AI With Samraj Matharu on the Six Sells Podcast

<p><strong>In this episode of The Six Sells Podcast, Mike Nicholson and Fern Potter spoke to Samraj Matharu about ‘Responsible AI’.</strong></p><p><br></p><p>Samraj is the Founder at The AI Lyceum which brings together a community of global practitioners and experts from technology, media, policy, healthcare, and academia to focus on ‘wisdom over intelligence’. </p><p><br></p><p><strong>Topics discussed include:</strong></p><p><br></p><p>•⁠ ⁠Samraj’s views on the importance on thinking ‘properly’ and ‘deeply’ with origins in ancient philosophy and its relevance to AI</p><p><br></p><p>•⁠ ⁠The relevance and importance of human judgment and using AI as an assistant and an accelerant, not a replacement</p><p><br></p><p>•⁠ ⁠Testing and learning to give clients a measurement on success and the psychological safety of growing in increments </p><p><br></p><p>•⁠ ⁠2026 as the year to test and fail cheaply </p><p><br></p><p>•⁠ ⁠Thinking deeply about philosophy and acting on it </p><p><br></p><p>•⁠ ⁠The danger of humans relying too much on AI and the importance of building an infrastructure that helps future generations </p><p><br></p><p>•⁠ ⁠Defining what good thinking actually is</p><p><br></p><p>•⁠ ⁠Memory, context and training data </p><p><br></p><p>•⁠ ⁠Personalised medical agents and the case for self analysis based on context </p><p><br></p><p>•⁠ ⁠Trust and AI and the growth in this mindset </p><p><br></p><p>•⁠ ⁠Tonality, ethics, consumer trust and responsibility </p><p><br></p><p>•⁠ ⁠How to build ethics, morals and principles as a standard</p><p><br></p><p>•⁠ ⁠The nuances in content when building a non-biased platform</p><p><br></p><p>•⁠ ⁠SEO strategy and ethical output</p><p><br></p><p>•⁠ ⁠Samraj’s view on who is doing ethics well in AI and agentic frameworks </p><p><br></p><p>•⁠ ⁠How brand safety will change in an AI world</p><p><br></p><p>•⁠ ⁠The challenge of translating the principles and values of the business into tech rules </p><p><br></p><p>•⁠ ⁠Deterministic and probabilistic outcomes and setting parameters for the agents to operate within</p><p><br></p><p>•⁠ ⁠The difference between algorithms and agents and protecting your brand and the consumer </p><p><br></p><p>•⁠ ⁠The echo chamber within LLM’s and the risk of bias and ‘people pleasing’</p><p><br></p><p>•⁠ ⁠The fingerprints in human writing, linguistic forensics and LLM’s</p><p><br></p><p>•⁠ ⁠The importance of transparency of the code and intellectual context</p><p><br></p><p>•⁠ ⁠Having consistent hum