Giggso Co-Founder on Responsible AI Governance at Scale

How can enterprises harness powerful AI responsibly? In this podcast hosted by Mphasis VP of Products Chenny Solaiyappan, Giggso Co-Founder Ravi Venugopal speaks on responsible AI governance at scale. He shares how governance, security, and observability form the guardrails for enterprise AI, why “AI must be farmed, not left to run wild,” and how leaders can balance experimentation, innovation, and compliance while upskilling their workforce in an AI‑native world.

Subscribe to the Product Talk podcast on Spotify and Apple Podcasts and catch every conversation with leading product executives.

---

---

LISTEN ON ITUNES

---

LISTEN ON SPOTIFY

---

---

Show Notes

1. AI must be governed, not left to run rogue – it should be “farmed” with clear boundaries.
2. Generative AI is inherently inconsistent, so it can’t be blindly trusted in production.
3. Human‑in‑the‑loop is essential for safety, trust, and accountability (like Tesla FSD).
4. Regulation and governance enable AI at scale; they’re not blockers to innovation.
5. Unmanaged AI can cause real harm, from psychological impact to abusive content.
6. Observability = monitoring AI behavior and drift, like watching a child’s patterns.
7. giggso started as conversational intelligence, then pivoted to AI for customer support.
8. Trinity is giggso’s enterprise platform for AI security, observability, and incident management.
9. AI red teaming as a service (AirTask) helps small teams harden agents with firewalls, monitoring, and simulated attacks.
10. Two main customer segments: very large enterprises (Trinity) and SMBs/agent builders (AirTask).
11. Assist vs. augment vs. automate: repeatable SOPs → assist; judgment‑heavy tasks → augment; end‑to‑end agents → automate.
12. Fully agentic systems exist but still need oversight; accuracy gaps (e.g., 30% misclassification) remain.
13. AI is great for 0 → 1, but humans are still key for taking results from 1 → 10.
14. Continuous upskilling is mandatory; people are replaced by others who use AI, not by AI alone.
15. Use AI tools to learn faster (e.g., notebook‑style LLMs, Perplexity, curated feeds).
16. At giggso, learning is formalized via certifications, university programs, and leadership leading by example.
17. Domain experts must also embrace AI, not rely solely on legacy experience.
18. Shift from output to outcome – what matters is achieving results within guardrails, not activity logs.
19. Regulation is how we “farm” AI into reliable, high‑value systems rather than wild experiments.
20. Ethical AI is a human responsibility – powerful tools must be used for societal benefit, mindful of real energy and water costs.