If we haven’t reached this stage already, the 3D engineering design business is rapidly approaching a tipping point. Those engineering companies who haven’t yet adopted AI will soon become uncompetitive.
But not all is perfect in the realm of Artificial Intelligence.
However, let’s start with the positives.
We all know the design phase for any new product, service or process is expensive. Therefore, we might argue, saving money is perhaps the most critical advantage of adopting AI in Engineering Animation. We’re all told AI will allow engineering companies to do more work in less time with less resources.
Marcin Frąckiewicz argues generative AI design software produces multiple options based on different inputs and conditions such as weight, strength and material constraints. Additionally, AI-powered simulations predict how an Engineering Design will likely perform under stress, heat, and vibrating conditions.
All good so far.
Spencer Chin quotes a Forrester Research survey of 163 engineering managers, the majority of whom believe not having AI solutions to perform laborious tasks will result in a loss of market share. Presumably, the time saved will allow engineers to focus on more complex and creative tasks that require human expertise and decision making.
Walter Shields uses terms like “massive transformation” and “game changer” to explain his perspective. Shields makes a valid argument that AI requires access to vast datasets to produce entire 3D Engineering models that are optimized for specific performance metrics. AI can help automate time consuming tasks like site planning, resource allocation and scheduling. According to Shields, all these benefits will help to reduce costs, improve efficiency, delivering construction projects on time and on budget.
I could go on and quote many other sources. By now, you’re getting the picture. No one it seems, has a bad word to say about the use of AI in 3D Engineering Design.
But let’s remember, AI can only generate outcomes from existing data. So, aren’t we simply getting a regeneration, or maybe a reorganization of previous material in a different form whenever we use AI? Aren’t we just shuffling the (many) cards, to give us new combinations?
Jennifer Chu quotes an academic study done at MIT about a subset of AI called Deep Generative Models (DGMs). Apparently, a DGM is a broad term that can be applied to any machine-learning model that is trained to learn distribution of data and then use that to generate new, statistically similar content.
The MIT study revealed pitfalls of DGMs when tasked with solving 3D Engineering Design problems of bicycle frame design. Apparently, the AI models mimicked previous designs well enough but faltered on engineering performance and requirements.
However, this is not all bad news. When a DGM specifically created with engineering-focused objectives, rather than based on statistical similarities, was tasked with a bicycle frame problem, it produced more impressive and higher-preforming frames.
You get the idea, right? The more specific the dataset provided to solve a problem, the better the outcome. Which leads to the next potential problem with AI; the dangers of “techno-solutionism”. Put simply this term means a reliance on AI as a solution rather than a tool. For example, if a designer or engineer is out of ideas to solve a problem, throwing a dataset of faulty designs into an AI engine will only produce alternative designs with the same, inherent problems.
Cremer, Bianzino, and Falk offer intriguing perspectives about the potential for generative AI to disrupt creative work. The authors concede a tsunami of AI assisted creative work is on its way to our customers and all companies should be well prepared for this.
One argument that intrigued me was the potential for a “techlash” whereby consumers and customers place more value on authentically created, “human” creativity than algorithmically generated content. Consequently, the authors argue, our stakeholders may rely more on trusted human sources rather than machine-generated information.
So, one perspective that may be drawn at the beginning of the AI revolution is; use Ai for the grunt work, not the creative work. Thus far, no technology has been proven to surpass our imagination.