AI at Hyperspeed and HyperScale

Creating the roadmap for change in a surreal AI world

We were astounded by Nathan Benaich’s latest report on the State of AI in 2023. Not so much by the pace of change, but by the breadth of change. We recommend you take a few minutes and read through the report. We think you’ll agree that while the hyperspeed of the AI market is formidable, the hyperscale is inspiring. In other words, what part of your business can AI reach that you haven’t considered before?

Here are a few things that caught our attention:

• Large Language models are the new architecture for innovation, touching everything from medicine and pharma to music, education, and agriculture. Wherever there is data, there is opportunity to harness new processes for driving workflows and outcomes. But, of course, the quality of content governs.
• This has been a breakout year in image, video, coding, voice, and animation. The chip makers like NVIDIA are making the money, but profits are expected to migrate to other corners of the AI market as high functioning applications are released.
• Regulations have been slow to evolve, probably due to the absolute challenge of grasping the direction and risks of AI technology. Is there a genuine safety risk to society, and if so, how do we measure it exactly? How would risks be mitiagted with regulations without creating constitutional conflicts? There is much more work to be do in this space.
• In the world of software engineering, the impact of GenAI has been decisive. The report cites studies on the productivity gains of software development teams through adoption of products like GPT-4, AlphaZero, WizardLM and CodeLLaMA. We have noted the impact of LLMs on engineering productivity and quality in our prior posts, but the data cited in this report shows that the observations are more than anecdotal. The integration of LLMs into the workflow of software delivery from design to testing is more than a preference. For the best software shops, it is a requirement.

One captivating part of the report are case studies where LLMs are learning software tools, planning projects and reasoning about next best actions. This is a whole new level of innovation, and impacts workflow beyond the world of software engineering. A fair question to ask is how much of this hyperspeed is hype? But the implications are too significant to ignore.

Of course, not every product or use case cited in the report is ready for production. But our takeaway is that when the Industry is moving at hyperspeed, you have to be hypervigilant about opportunities and risks, and prepared to 'pivot and scale' with AI product where appropriate. Our experience is that prototyping is a highly effective way of testing the value, avoiding the hype, and identifying the risks of adopting AI in your core processes. And with these proof points in hand, we suspect you’ll move at hyperdrive to scale your innovations.

Give us a call. We are a leading AI prototyping firm, with the latest methods, components and experience to help you quickly test out ideas. We can help you map out your roadmap for thoughtful change in this hyper AI market.