Extending Product Roadmaps with Generative AI: Acrolinx CEO Volker Smid (Part 6)

Sramana Mitra: It seems like Microsoft, AWS, Google will create sophisticated AI-optimized cloud environments. These environments are going to be so advanced that you would be hard-pressed to argue against using one of those.

Volker Smid: The other aspect is if you have the system on-premise, it’s hard to be connected to external services. All the advantage of this connected infrastructure goes away if you deploy on-premise. There’s still sensitivity and concerns around security of cloud. There are European customers who don’t want to use a service that is hosted by companies that sit in the US because of information security.

It’s not straightforward, but, over time, everybody will realize that you will get a lot more security if you leverage the public cloud infrastructure.

Sramana Mitra: If generative AI were in the picture when you were starting to build this product, would anything have been different?

Volker Smid: I think so. We talked about efficiency earlier. The question now becomes, ”Where’s the human interaction?” I still believe that if you create content for a responsible enterprise, you need to have a human in the loop. We will quickly come to a point where you can create a manual automatically using several data pools. If you want to create a manual for software, all you need is access to the source code, wire frames, provide this to a generative AI solution so they can figure out what the user interface means from a user’s perspective.

These services will come over time. If we had generative AI available earlier, we would be at the point to have customers gain more efficiency to create manuals automatically. We will see an astonishing amount of possibilities coming our way. We will always have to decide what to do with it and where’s the limit of efficiency gain. Now is the time to build the capability for the enterprise.

Sramana Mitra: Put yourself at the 30,000 foot level and think of a new entrepreneur who has a decent amount of AI knowledge, what are some product ideas that come to you as possiblities of where a new entrepreneur can build companies.

Volker Smid: Let me connect to what I just said. Let’s consider An entrepreneur that understands the complexity of mechanical engineer. When a mechanical engineer creates a new physical product using a CAD, if you combine that with generative AI while the engineer creates a physical product, an entrepreneur that connects all the dots can use the data to automatically create every use manual in every possible language in every possible aspect and regulation.

Where’s the entrepreneur that understands that this is the data pool? This is the data pool in the PLM side. This will help me tune a model to help me create a manual on the fly. This will be a successful entrepreneur. The time to market advantage that you give to a company that creates a new product is vast. Normally the user interface is the last piece that is being created.

Sramana Mitra: Do you have other ideas?

Volker Smid: Do the same for the software environment. Software is a product that people pay money for. The problem with software is, in today’s world, you create a new version almost every day. The problem is, the documentation is always late. Normally it doesn’t exist. A great entrepreneur that understands the whole tooling environment will do great.

Sramana Mitra: Your ideas are all highlighting both the problems and opportunities in documentation and user manuals of different types of systems. Mechanical engineering system is the world of CAD and PLM. The software engineering world is the world of GitHub and open source, and APIs.

Documentation, given all of that, is complicated. You are saying that generative AI can be automated. An entrepreneur can build products to achieve these kinds of outcomes.

Volker Smid: Yes. Take the legal system. In the US, almost every documentation in the legal system is publicly available. One of the things that I learned in the US is that the average price for a lawyer is $750 an hour. With generative AI and law in general, we should take all this and create a specialized model. It will create an industry below the high-paid legal system that probably is able to do 80% of the work. This will create another industry. In five years from now, I think there will be a complete new industry.

Sramana Mitra: That’s why I ask these questions – to stir up ideas. It was a pleasure. Thank you for your time.

This segment is part 6 in the series : Extending Product Roadmaps with Generative AI: Acrolinx CEO Volker Smid
1 2 3 4 5 6