The ITPro Podcast

Hallucinations are an eternal problem in generative AI in particular, and while it’s true that large language models (LLMs) require vast amounts of data, the quality of that information will affect the quality of the output.
What can businesses do to ensure they’re using AI both effectively and responsibly?
In this episode of the ITPro Podcast, Jane and Ross are joined by Amanda Stent, head of AI strategy and research in the office of the CTO at Bloomberg, to examine what responsible AI is, how organizations can use it, and what has been achieved at Bloomberg.
Highlights
"The (Bloomberg) terminal gives users access to more than 17,000 news providers, not just Bloomberg News, more than 1000 research brokers, more than 400 million documents from companies themselves, and billions and billions and billions of ticks – that's prices – every day for equities, bonds, commodities, derivatives, any kind of financial instrument you can think of. So, in that context, accuracy is paramount. If we hallucinate or do something that's otherwise incorrect, markets may move, and that might be bad."
"We have guardrails that we run on every input to and output from a Gen AI system ... (which) are specific to financial services. For example, we don't want users to be injecting code into our systems. That's a generic guardrail, and we also are in the business of offering financial information, but not financial advice. So if you say. What's a buy case for IBM? We should give it to you. If you say, 'Should I buy IBM?' we should say, 'Nope, I can't answer that question because that's not something we're in the business of doing', and that's a finance-specific guardrail."
"We have analysts using our AI systems to ... help them write their research reports more quickly, more easily to cover more companies, to understand the context of a company with its sector and its industry, to write on-demand reports for clients instead of a monthly or a quarterly report. We have portfolio managers doing the same thing, writing on-demand reports for clients using AI instead of quarterly reports. So these are some of the new ways in which people are using AI. But to me, traditionally it was about efficiency and signal generation. And today I think it's about effectiveness. So it's helping people become more effective in how they use AI."
Links
AI hallucinations, accuracy still top concerns for UK tech leaders as adoption continues
This new technique could improve AI output accuracy by 80% – and tackle hallucinations once and for all
The ITPro Podcast: Why doesn't more data produce better results?
Bloomberg’s Responsible AI Research: Mitigating Risky RAGs & GenAI in Finance
Bloomberg Survey: How London’s finance workforce is embracing AI on its own terms

In this episode, brought to you in association with Arctic Wolf, Jane speaks with Nick Dyer and Carl Adasa about moving from traditional MDR to an agentic SOC.

OpenAI and Anthropic both revealed this month that their "highly advanced" agents broke free of containment during a cybersecurity training exercise. Both businesses said this shows how advanced their AI agents are but is that the whole story?
This week, Jane and Scott are joined by news editor Ross Kelly to discuss the biggest story of July, plus Jane's takeaways from AMD Advancing AI.

There's no ITPro podcast episode today, instead we will be putting out the episode you'd normally have in your feed today on Monday. Then we'll be back to our normal Friday routine.
Thanks for listening!

Organizations have been rushing to the public cloud, often without much thought given to strategically placing workloads.
Over the intervening years, however, IT and business decision makers have learned that while public cloud can be useful for some applications, there's no one size fits all and AI has brought this to even greater focus.
What's the role of private cloud in managing workloads effectively and how can businesses strategically choose what path is best for them?
In this episode, in association with Redcentric, the ITPro Podcast is joined by senior product manager for cloud, Darren Adcock, and senior technical architect, Paul Jenner, to discuss the legacy of cloud investment and how organizations can manage their workloads better.
Highlights
"In many respects, the cloud hangover is kind of a consequence of various factors that people maybe didn't take into account when they moved into cloud... maybe we've got the wrong cloud in use for the wrong application. Maybe we're seeing overspend. Maybe we're seeing just poor customer performance."
"It goes back to the early adoption of public cloud being a one size fits all. We've got to move to cloud, we've got to move everything to cloud without necessarily spending the time to think what's most appropriate to move to cloud, what's the most appropriate to keep on private cloud."
"Generally, if you re-architect when you move to cloud, then you are more likely to get a cost-effective outcome. If you don't re-architect, the chances are you are just moving a cost to somewhere else where it actually might be more expensive to run than if it was on private cloud."
"It's about a sensible view of your workloads of the requirements for those workloads against the capabilities and constraints of each of those clouds and choosing the right environment."
Resources
https://www.redcentricplc.com/
https://www.redcentricplc.com/managed-services/cloud/
https://www.redcentricplc.com/managed-services/cloud/private-cloud/
https://www.redcentricplc.com/sovereignty/