The Automotive Industry’s 4th Disruptor: Agentic AI in the Enterprise
- Alex Oyler
- 5 days ago
- 5 min read
Updated: 1 day ago
“We are witnessing an explosion with agentic AI,” AWS CEO Matt Garman observed this week during his keynote at the company’s marquee annual conference, AWS re:Invent. “Every single customer experience, every single company, and frankly, every single industry is in the process right now of being reinvented.”
Call me naïve, but coming into an AWS conference, I didn’t expect the entire two-hour keynote from their CEO- plus the second-day keynote from their VP of Agentic AI, Swami Sivasubramanian – to focus on agentic AI. I thought I’d be hearing about traditional enterprise infrastructure and platform enablers like analytics applications, container management, serverless services, and so on – but no, just AI.
The strategy worked. You can now consider this fence-rider fully AI-indoctrinated. But I’m not sure that you could say that for the rest of the automotive industry yet.
It’s an industry being disrupted thrice over by software-defined vehicles, electrification, and autonomy, so you’ll have to forgive industry leaders, particularly those in organizations with distributed decision-making authority, for struggling to embrace agentic AI. It might feel like adding agentic AI as the fourth axis of automotive reinvention might stretch R&D budgets to the absolute limit – much less attention spans.
But the agentic AI reinvention has a distinct uniqueness to it that the others do not. SDVs, EVs, and AVs enable customer value. These are the table stakes that automakers must invest in and manage in order to protect their brand, their relationship with customers, and positioning within the market. This calls for accelerated R&D efforts, capex investments, and restructuring efforts that reposition skills and supply chain partners to deliver on the promise of customer value.
AI, however, has a fundamentally different value proposition.
It’s Not A Chatbot, It’s a Minion
“AI assistants are starting to give way to AI gents that can perform tasks and automate on your behalf…this is where you’re starting to see the returns that match up with the promise of AI,” Garman continued in his keynote address.
The first frontier of generative AI in automotive was the in-vehicle voice assistant, which made sense. When ChatGPT was released to the public in November of 2022, the potential for the technology to revolutionize conversational voice assistants became immediately clear. This value proposition extended to “copilot”-like applications that offer fast, natural-language access to information that helped people get done things faster. While not without its warts, the potential of the technology led to companies like Mercedes-Benz launching early versions of ChatGPT into their vehicles just one year later.
Agentic AI is different. Yes, these are still trained large language models, but through breakneck progress across the AI ecosystem – from Nvidia’s hardware to Anthropic, Meta, OpenAI, Google and others’ models and extreme capital investment in computing capacity, context windows expanded rapidly. This meant that AI companies could train their models with more & more data, leading to more accurate, robust, and specific outputs. By implementing agents that encapsulate specific context windows within an enterprise environments, companies can now effectively replace monolithic applications with agents – the end-state for serverless computing architecture.
By creating agents that orchestrated workflows, managing inputs & outputs autonomously, CIOs all of a sudden saw a future where instead of developing and maintaining hordes of legacy applications, their developers could build agentic architectures that managed, improved, and triaged in real-time.
This is the new frontier of agentic AI both for the automotive industry and for all industries – not in the car, but in the cloud.
Now is the Time for Automotive Leaders to Invest in Agentic AI
In front of a capacity crowd, leaders from Toyota Motor North America’s Enterprise AI and Toyota Connected organizations detailed their agentic AI journey, illustrating the rapid pace of innovation and optimization. Under the leadership of key automotive industry AI voices like Brian Kursar and Dave Tsai, Toyota has developed a template for how other automotive companies could successfully enable agentic AI across the enterprise.
Toyota’s ability to deliver on this vision has been many years in the making. By building and delivering on a strategy that prioritized in-house AI and engineering talent development, Toyota sits in a position of envy amongst many of its competitors – one that can lean on its skilled, diverse, and energized workforce to lean into the rapid pace of innovation amongst AI tooling leaders like AWS and its competitors.
This week, AWS launched a number of new agentic AI services aimed to enterprises with all levels of engineering capability, with some of the newest services – like Kiro – enabling spec-driven development. This lowered barrier of entry means that automakers and suppliers that face increasing resource constraints and competitive pressure can quickly and easily lean into tools whose first and foremost value is allowing faster product development with better quality.
This is the fundamental yet surprisingly quiet promise of agentic AI for the automotive industry. It’s not a new powertrain, supply chain, vehicle technology platform, or means of mobility. Rather, it’s a superpower that allows automakers to respond to those disruptive pressures more efficiently and quickly, all the while enabling new pathways for innovation across the company.
Where Automotive Agentic AI Goes From Here
Multiple frontiers of innovation are emerging within the automotive industry as AI transforms what IT teams can do. Three that emerged from re:Invent this year include:
SDV Complexity – The biggest thorn in automaker SDV strategies is managing a complicated, fragmented mess of embedded software across architectures, regions, and even internal silos – not to mention software suppliers. At re:Invent, General Motors demonstrated how its been able to refactor legacy code using agentic AI, resulting in tangible improvements to cyclomatic complexity while maintaining adherence to standards like MISRA C.
Industrial AI – Building digital twins of factory floors and supply chain, using agents to monitor operations, predict failures or downtime, triage issues, and optimize (or even support refactoring of) business processes that enable more efficient, reliable factory operations.
Managing All That Vehicle Data – With automakers being sensitive to any and all compute headroom built into cars, on top of the long operational lifespan of these assets, running agents in the car in the near-term will be a challenge. But as OEMs deliver on SDV strategies and edge inference capabilities increasingly optimize, the possibility of agents running in the vehicle environment to support various customer and OEM data-intensive applications becomes increasingly realistic, improving customer privacy, lowering the cost of regulatory compliance, and enabling more predictive and personalized in-vehicle customer experiences.
Security AI – 2025 was a banner year – in a bad way – for automakers suffering from various cybersecurity challenges at the hands of black-hat hackers. Furthermore, vehicles are some of the hardest assets to secure due to their long lifespan – which often leaves them with deprecated cryptographic capabilities just a few years into a program’s run. Agentic AI-enabled security agents in the cloud and in the car could allow automakers to more efficiently take a much more proactive stance to cybersecurity. Although industry leaders continue to invest and protect their businesses and products today, the scalability of agentic AI could improve the resiliency of business operations many times over.
There’s no question 2026 will bring more progress, and we’ll be keeping a close eye on how innovations in the AI space could affect automakers and their partners. In the meantime, we’re excited to continue working with our customers to learn from industry best practice, apply AI technology meaningfully and prudently, and identify the right partners that help them achieve true AI return on investment.
Next up? We’re back in Las Vegas for CES 2026. To set up time with our leadership team and analysts email us below:
To stay up-to-date on all the latest news about in-vehicle technologies, subscribe to our newsletter!