In an exclusive interview with Cyber Intelligence, Mike Finley, the Co-Founder and CTO of AnswerRocket, a business intelligence platform that deals with big data and AI agents, explains what generative AI can do for companies right now.
Cyber Intelligence: Where are we right now regarding the successful adoption of generative artificial intelligence (AI)?
Mike Finley: AI is changing faster than people are capable of understanding. So the general misunderstanding of what AI can do is going to be a lasting problem. The fact is that key scientists believe AI is now capable of improving itself, meaning we are at the start of a runaway path forward. At AnswerRocket, our basic DNA is artificial intelligence (AI) to enable business intelligence (BI). This obviously took a new direction with the widespread introduction of generative AI, but our basic approach remains the same.
Cyber Intelligence: How can small or medium-sized organizations take advantage of generative AI while avoiding the pitfalls of what is obviously a fairly nascent technology?
Mike Finley: One of the ways of adopting early forms of generative AI to even a relatively small organization’s benefit is to use what we call “evergreen” uses for AI that can benefit everyone in an organization. The best way to instigate AI is to start training AI models with examples of what your company wishes to achieve. This is a tremendous opportunity for SMEs.
Cyber Intelligence: Can you give a practical example of ways in which a medium-sized organization can use generative AI to its advantage right now?
Mike Finley: Microservices are a good example. For example, product returns are a microservice. Merely instigating chatbots is not really a workable strategy. If generative AI is to be genuinely useful then it must replace human decision-making at a simplistic level, such as product returns, but guardrails must be firmly in place for this to work effectively. If the company has a 90-day return policy, then AI can be instructed to follow this rule. There are also seemingly little things that AI is easily capable of doing right now that can make a tremendous difference in terms of streamlining time and resources. One obvious one is ensuring that pictures are in focus. For example, a dissatisfied customer may have uploaded a picture of a faulty product. AI can then inform the customer if the picture they have sent is in focus or not.
Cyber Intelligence: Are there any examples of ‘evergreen’ applications of generative AI today?
Mike Finley: The Oak Ridge National Laboratory in the US is already using drones to investigate incidents such as a power outage after a storm. Sending a human team could cost $1,500 per incident, whereas drones can perform the same checks for around $25. The key to being an efficient early adopter of generative AI is to start small but not slow.
Cyber Intelligence: Can you think of any more uses for generative AI that could be applied more generally?
Mike Finley: Computer programming is a key example. Using highly paid IT staff to carry out simple programming tasks is analogous to hiring a stone mason to carve gargoyles onto Notre Dame. Today, we would cast the whole lot in concrete. In a similar way, AI can already be used to carry out straightforward programming tasks.
Cyber Intelligence: What do think the new incoming US administration in Washington will mean for the rollout of generative AI in 2025? While the incoming president has proven business skills and political acumen, it seems that billionaire entrepreneur Elon Musk will be consulted regarding technological issues such as AI. How will this work in practice?
Mike Finley: I think it will be interesting to see how Musk’s AI offering, Grok, handles some of the challenges facing generative AI as I think it is set to be one of the strongest players.
Cyber Intelligence: What kind of challenges are those?
Mike Finley: There is a growing view that generative AI will have to teach itself how to be morally good in a human sense. The danger in telling it that some areas are bad and, therefore, off limits is that AI will become unduly curious about these off-limits hidden corners or knowledge. The new approach is that, by not teaching it to be good, it can learn to be good.
Cyber Intelligence: Thank you.
