AI Australia Podcast

We change things up in today's episode and grill two Mantel Leaders, Samuel Irvine Casey and Vihan Patel, who have been in the trenches over the past 12 months building AI and Agentic Solutions. They help us unpack 2025 and predict what 2026 will bring including:
The ROI Reality Check: As the hype of 2025 settles, 2026 is framed as the year of accountability, where organisations must prove the economic value of their AI deployments.

The Token Economy: we deep dive into the sustainability of AI costs, questioning whether Silicon Valley will continue to subsidise token prices or if costs will spike as models move into high-scale production.

Knowledge Collapse & Ethics: We explore the risks of "AI feeding on AI," the potential loss of critical thinking as we increasingly rely on AI generated knowledge, and the importance of new Australian legislation regarding AI traceability.

The AGI Debate: we question how imminent "Artificial General Intelligence," including how current progress is driven more by clever engineering and high token usage than a fundamental breakthrough in machine consciousness.

If you want to hear more about Samuel's lessons learnt from building Agentic Systems take a look at his blog from 2025: https://mantelgroup.com.au/taking-agents-to-production-our-agentic-ai-approach/

Forget building dashboards! The CEO of ThoughtSpot reveals how conversational AI is killing traditional business intelligence and putting a personal, 24/7 data analyst on every employee's desktop. Are you ready to fire your dashboard and start talking to your data and can you trust the answer when you do?
In this episode, we talk to Ketan Karkhanis the CEO of Thoughtspot, a former Salesforce and Cisco executive, who explains that "AI is the new BI," arguing that businesses should stop building dashboards and start having conversations with their data. He introduces ThoughtSpot's AI agent, which acts as a personal analyst that provides actionable insights (what, why, and what to do next), with a new "A-Team" of specialised BI agents coming soon.
To address trust issues, we explore who you can use deterministic analytics and transparency to combat AI hallucinations and secure enterprise adoption. Ultimately, Keten stresses that successful AI integration—like automating data modeling or transforming his own staff meetings—must be tied to clear corporate ROI and a culture that empowers change across the organization.
Read Mantel's recent take on Natural Lanugage insight tools: https://mantelgroup.com.au/dashboards-natual-language-insights/

Discover how Swimming Australia is diving deep into data and AI to secure Olympic gold!
Today we chat to Jess Coronas, Head of Performance Insights at Swimming Australia, and discuss the digital transformation journey of Swimming Australia. From the initial ideation and data driven insights to the development of the groundbreaking Training Insights platform, powered by cutting edge computer vision techniques. Learn why human-in-the-loop AI is crucial for coach buy-in, how data creates a competitive edge, and the future vision for scaling analytics from the 50m sprint to the 1500m race. It's a masterclass in applying design thinking and marginal gains to a high-stakes, world-class sporting environment.
Learn more about Mantel and AWS partnering to support Swimming Australia and their Training Insights platform:

What does factory floorplan design during the Industrial Revolution have to do with re-imagining user experiences in the ages of AI?
In today's episode, we chat to Dr. Haijun Xia, an assistant professor at UC San Diego and Director of the Foundation Interface Lab, discussing the intersection of Artificial Intelligence (AI) and Human-Computer Interaction (HCI). Dr. Xia, whose work focuses on developing foundational interfaces for the future information environment, explains HCI in simple terms and outlines the criteria for a good user experience, emphasising the need for interfaces that leverage and amplify human cognitive abilities while introducing new capabilities smoothly.
We discuss the shift from "one-size-fits-all" interfaces to adaptive and personalised experiences driven by AI and how large language models, allow users to interact at a higher level, simply stating their problem instead of manually performing low-level system manipulations. The conversation delves into the future of product design, including the prospect of personal AI agents that interact with software via APIs on the user's behalf, creating a "digital version of ourselves" and influencing how new products must be designed for both humans and agents. 
To find out more about Mantel's partnership with Dr Haijun Xia and download our recent white paper: https://mantelgroup.com.au/mantel-partners-with-ux-leader-and-expert-dr-haijun-xia-to-advance-next-generation-ai-experiences/
Want more detail on Mantel's Exec AI Literacy, visit our website: https://mantelgroup.com.au/ai-executive-literacy-program/

Join us for a conversation with Sharryn Napier, Vice President for APAC at GitHub, as she discusses leading the company's strategic direction and business growth across the region. Sharryn talks to GitHub's evolution from a code storage platform to the world's leading platform for building software with AI, now supporting over 150 million developers.
The episode explores the democratising power of AI in software development, through tools like GitHub Copilot, which lowers the barrier to entry (even for non-coders) and addresses challenges like the language barrier in coding for non-native English speakers. We discuss how enterprise customers, like the Commonwealth Bank of Australia (CBA), are leveraging AI coding agents to manage and refactor significant legacy tech debt, seeing projects completed in a fraction of the time. She highlights that the "risk of not moving fast enough" with AI is now a greater concern than the risk of adoption.
The discussion also delves into regional challenges, particularly Australia's flatlining productivity and the tech skills gap. Sharon advocates for AI as a mechanism to work "smarter" and compete globally. She also addresses the future of the developer role, viewing AI as a tool for augmentation, not replacement, that handles tedious tasks, increasing developer happiness, creativity, and shifting the role toward that of an orchestrator.