The Data Cost Dilemma:

Cut AI Token & Compute Costs in Half. Maximise Capabilities. Minimise Spend.

Wednesday 11, November | 12:30-3:30pm | The Angler, South Place Hotel

Lunch with Strategy

As organizations scale their enterprise AI initiatives, data leaders face an escalating challenge: exponentially rising token and compute costs threaten to cap innovation and erode ROI.

Join an exclusive group of Data, Analytics, and AI leaders for an intimate luncheon to tackle The Data Cost Dilemma head-on. Discover strategic, actionable approaches to drastically curb infrastructure overhead while boosting the precision, power, and scale of your enterprise AI workflows.

Enjoy fine food, great company, and candid conversations with executive peers who are all working to make their data stacks and AI investment produce maximum value without blowing the budget.

What to Expect

The Strategy team will frame each discussion, while the conversation is driven by the attendees. The format is designed to encourage discussion across IT, data, AI, risk, and business leadership.

One-table executive lunch held under Chatham House Rule. There will be no presentation deck and no recording.

Who is Attending?

Senior leaders from across the FSI sector, including banking, insurance and wealth management.

This is a deliberately intimate event, with a maximum of two attendees per organisation, representing platform, data, and risk leadership.

About Luncheon:

Where: The Angler, South Place Hotel | 3 South Pl, London EC2M 2AF

When: Wednesday, 11 November 2026, 12:30pm – 3:30pm

Guided Discussions

Why is our AI spend climbing faster than the value it returns?

AI pilots are multiplying, cloud costs continue to rise, and many production AI systems still fail to deliver consistent accuracy. Is the solution simply larger models, or is the answer found deeper within the AI stack?

Are we building applications we'll need to rewrite in 12 months?

AI models and pricing change fast, yet many organisations build around whatever model was leading at the time. What architecture will outlast the next three generations of models and who should own that call?

Can we prove what our AI did, with whose data, and under which rules?

One in eight employee prompts contains sensitive data, yet many organisations still lack a full audit trail. With DORA, NIS2, and the EU AI Act now in force, how ready are they to prove compliance, governance, and accountability under audit?