2023 has been the year of AI. In the wake of ChatGPT’s debut at the end of last year, we’ve seen a rush of AI-powered startups burst onto the scene and a wave of investors eager to find the next OpenAI. All the hype around generative AI has led to much speculation around its potential applications, from education, to healthcare, to climate.
As a climate software investor focused on digital solutions that can help scale the energy transition, our team at Energize has been evaluating AI-powered climate technology for years – before “AI” became the term of choice. We’re continually analyzing the risks, the hype and the impact potential of AI in the climate sector through our proprietary research and our relationships throughout the ecosystem.
At Energize, we don’t believe in AI for the sake of AI. We’ve been outspoken in our belief that hype cycles don’t drive sales cycles, and we remain laser-focused on solving real customer pain points. We also believe that AI in its current form poses risks around accuracy, equity and data privacy, which need to be better addressed and governed. That said, we do believe AI can deliver real value for customers within the climate space. We’re already seeing AI and machine learning (ML) models crunch down soft costs – by as much as 90% in some of our investments. Rather than viewing AI as a buzzy technology that needs to be chased down, we see it as a useful tool in the customer success toolbox that, when powered by high-quality data sets, can play a significant role in addressing customer pain points in climate.
Furthermore, while applications in generative AI have exploded in recent months, the use of AI technology in climate software startups is nothing new. More than half of Energize’s portfolio is AI/ML-native, while the majority of the remainder of our portfolio companies have incorporated AI to enhance their process flows or leverage large proprietary data sets. They’ve been using AI to provide real climate solutions ranging from forecasting electricity demand to automating solar design to creating dynamic climate risk models. Their solutions exemplify Energize’s investment thesis, which has been guided by the idea that better data can lead to asset owners making better decisions about things that can be difficult for humans to grasp – i.e., many challenges in the energy, sustainability and industrial spaces. Moreover, we believe climate software companies are best positioned to utilize AI to capture value and serve customers, as opposed to newer entrants that do not have the volume nor quality of data needed to generate accurate outputs with AI.
We decided to pose the topic of AI’s implications for the climate sector to an expert panel of climate tech CEOs and a venture investor at our recent event, Energize NEXT: The Climate Software Summit, earlier this month. Panelists included Bilal Zuberi, General Partner at Lux Capital; Reza Zadeh, CEO at Matroid; Stephan Rohr, Founder and Co-CEO at TWAICE; and Helena Merk, Co-Founder and CEO at Streamline Climate.
They delved into their own experiences and points of view on climate and AI, from what customers are saying to implications of AI technology to look out for. We’ve synthesized their insights into three major takeaways:
Takeaway #1: English is the new programming language.
The growing sophistication and popularity of generative AI tools is bringing about a broader transformation in the way we design and access cutting edge technology. “The technologies behind machine translation have gone consumer. Suddenly, what was at the hands of ML engineers is in the hands of everyone in the world,” explained Reza.
With this shift comes the potential for fundamental changes in customer expectations and how climate software companies structure their businesses. In the same way the internet and subsequently smart phones gave digital-first and mobile-first companies like Amazon and Instagram massive competitive advantages, the accessibility of AI means that tech companies that don’t utilize it risk being left behind quickly. “Pretty much every company should be using ML and AI,” he summarized.
Takeaway #2: Solving customer problems still matters the most.
However, when asked what resonates most with climate tech customers, the CEOs’ responses were resounding: Customers still care more about the solution than the underlying technology. No AI tool eclipses the importance of climate tech companies understanding their customers’ pain points and developing solutions that effectively address them. And climate tech companies should act accordingly when marketing themselves to current and prospective customers.
For Matroid, whose computer vision software enables customers to eliminate human error in manufacturing machinery, customers care most about detecting defects and incidents, not the fact that Matroid’s platform is AI-first. For TWAICE, a predictive battery analytics platform that uses AI, customers care most about their ability to make reliable, data-driven decisions to improve profitability, operations and battery lifetime. “In conservative industries, it is not always beneficial to apply an AI-centered communication,” explained Stephan. “Rather, we focus on customer value.”
Takeaway #3: Customer attitudes towards AI are complex.
When it comes to customer sentiment towards AI, there isn’t a hard and fast rule. For example, many climate tech companies serve customers in sectors where data privacy concerns are paramount, such as finance and critical infrastructure. For these customers, buying a tech solution that utilizes AI can feel more like a risk than a selling point. “Many of our customers who come from a deeply scientific background – AI is concerning to them. They are very worried about their privacy,” Helena explained.
On the other side of the spectrum, some customers are seeking out AI – sometimes just for the sake of AI. “Enterprises today are looking to incorporate AI because their shareholders expect it, not just because they want to use it to cut costs,” Bilal said. The bottom line? Climate tech companies must understand who their specific customers are, including their motivations, their concerns and their stakeholders. Customers seeking out AI should be able to easily find it in a climate company’s messaging, while customers who are hesitant must be able to have their concerns addressed thoroughly.
The most exciting part of the AI hype cycle is that it’s happening in real time, and these takes will continue to evolve alongside the technology. We look forward to continuing to foster these conversations throughout our network, and we’re excited about the potential for AI to increase the pace of the energy transition.