As with every new technology much buzz is generated, often motivated by players trying to bank on the fad for a quick sale, leaving some organizations to fall for the ever-present fear of missing out, only to later realize that it just wasn’t the right fit for their challenges.
Generative AI is no different and many are ready to call it everything from the equivalent of the advent of the Internet to the new Blockchain or a solution looking for problems to solve – I have heard so many arguments on both sides, from the most pragmatic to some bordering a level of paranoia that would make Stalin blush.
Instead, I’d rather be pragmatic: I have personally found the technology to be highly practical for many uses cases and I’d like to share a few of those we currently use daily at Phi Research plus one currently being built for a client.
As a B2B company, the most common goal for a main website chatbot is to generate leads. We set out to build it as a means for our prospects to get some sense of what we know and what we’re capable of. Yet we wanted to include a twist: let’s give something practical to our prospects. Bits of knowledge and our experience that they can use straight away.
This is a common use case for Retrieval Augmented Generation (RAG), or more plainly, a text-to-text generative AI tool that structures answers based on a pre-defined body of knowledge. Add to that some controls to reduce the chance of abuse, and a pinch of commercial swag to convince the user to set up a meeting with us.
As I write this, Hermes has been live for about 3 weeks and, though we’ve added some tweaks here and there, the results have been very positive. Both from customer and prospect feedback as well as its main purpose.
One common request we receive is for support in strategic ESG program design and ESG reporting – with all its associated nuances (data assessment and strategy, data engineering, BI development and automation, etc.).
The world of ESG is complex and unfortunately only one of our partners is well versed on the topic and his time is scarce, better spent addressing challenges brought in by our customers, rather than educating other team members – which is already hard to coordinate.
Instead, we have put together Gaia, which is another RAG implementation fed with the most common ESG frameworks and some case studies. As a result, the whole team has a practical ESG guide at hand to enhance its knowledge, which can always be supplemented by Jose Miguel, our resident ESG expert, greatly expanding his reach with little effort from his end.
Much in the same lines, a customer in the mining industry has the challenge of encouraging some machine operators to adhere to protocols for handling equipment performance exceptions. These machines are rather large and complex, and consequently, user manuals are huge and often hard to navigate.
On the other hand, operators’ motivation is to fulfill shift goals, usually resorting to tried-and-true methods to get the cranky line back up. What’s more, this experience is highly valuable and sought after. Taking a step back to find the manual – or worse logging on to a computer they never use at an office –, find the right section, go through a wall of text, all the while pressure mounts.
From the line manager and above, the priorities start to shift: expensive warranties, contractual obligations and preventing personnel injury tend to be higher up, so reliance on documentation and protocol becomes of outmost importance.
Our solution: have both sides meet in the middle and reduce friction to access this corporate knowledge. Using Generative AI as a UI front to all sorts of manuals, safety protocols and similar, we are building a conversational agent that can act a personalized engineering consultant for the operators.
There’s still much work to be done on this project, but preliminary results are encouraging, and all stakeholders remain much engaged on the initiative.
As with every new technology, beyond the opinions of supporters and detractors, it is always important to keep in mind the problems and use cases the technology is good for and above all, experiment and discover how it applies to your own organization.
The state of the art of Generative AI can only improve over time, so developing internal projects can greatly help your organization not only uncover its potential to your very own challenges, but more importantly, developing a culture that embraces change and can swiftly adjust to major shifts in industry paradigms, priming it for capturing more competitive advantages.
Join us on a unique journey where GenAI takes on the role of a graduate student, assisting a busy professional in a small company. Discover how GenAI becomes a collaborative partner, helping with tasks, generating ideas, and refining content. Explore the dynamic process of initiating the dialogue, venturing into uncharted territory, engaging in academic exchange, and crafting brilliance together. Uncover the highs and quirks of this innovative partnership.
Evolving business priorities demand periodic KPI updates. Reviewing and aligning KPIs with changing drivers of success and industry trends is crucial for accurate measurement and informed decisions.
Explore how Generative AI can revolutionize small and medium-sized businesses (SMBs) with three innovative use cases. Discover how SMBs can implement AI-driven solutions for customer knowledge giveaways, hyper-personalization, and channel consistency to gain a competitive edge in today's AI-driven landscape.