Cutting Auto Finance Costs with Generative AI

Here is what’s new in the AI world.

Generative AI Poised to Reduce Costs, Improve Efficiency in Auto Lending

A new report from McKinsey & Company highlights a powerful opportunity for your company. By leveraging Generative AI, you can significantly cut operating costs and boost efficiency across your operations.

The potential goes beyond simple customer service chatbots. For you, Gen AI can become a core part of your internal decision-making, generating insights and optimising processes at a scale that was previously impossible.

This is your chance to tackle long-standing inefficiencies and fundamentally improve your company's economics.

According to the report, Gen AI could reduce your cost-to-income ratio by 5 to 8 percentage points by lowering the operating costs that typically consume about 60% of your income.

The Key to Unlocking This: AI Agentic Systems

To capture this value, you should deploy "agentic systems." These are autonomous systems that use AI and machine learning to perform specific tasks. They are smarter than traditional automation because they can make dynamic, context-aware decisions and execute them independently.

McKinsey suggests you can integrate four specialised AI agent groups to transform your business model:

1. Your Remarketing Agent

• Let a Gen AI-assisted analyst help you choose the best channel and price for returned vehicles.

• Use dynamic forecasting to predict vehicle returns and intelligently allocate them across sales channels.

• Deploy an AI agent to continuously analyze market trends and generate commentary for your team.

2. Your Service and Operations Agent

• Empower your customer service with an AI agent that manages document workflows and coordinates repair bookings.

• Have AI and your human staff jointly analyze dealer performance, garage diligence, and insurance claims.

• Automate report generation for tasks that used to require heavy manual analysis, like identifying garage spend variability.

3. Your Procurement Agent

• Use AI to scan costs end-to-end and assess a vehicle's lifetime value.

• Automatically flag abnormal costs for your review, complete with recommended corrections.

• Automatically generate reconciliation reports for your team to review alongside the AI.

4. Your Sales and Pricing Agent

• Let AI agents calculate technical pricing (like residual values) and generate insights.

• Use a dynamic algorithm that suggests new monthly payments based on web-scraped market data.

• Have AI agents proactively reach out to customers with personalised commercial offers.

• Generate comprehensive reports to maintain transparency and control costs for maintenance, repairs, and insurance. These agents can flag abnormal insurance claims, identify unusual dealer spending, and detect irregular tire usage.

Where Should You Start?

The report recommends you begin by integrating AI into your remarketing efforts. Optimising your pricing across sales channels is both a high-impact activity and relatively straightforward to implement, allowing you to demonstrate quick wins.

By integrating these AI agent groups, you can create a more efficient, cost-effective operation. This doesn't replace your people; it empowers your human agents to act more quickly on improved insights and higher-value opportunities.

Xiaomi Takes on Tesla, Boston Dynamics with Open-Source AI Model

Chinese tech giant Xiaomi is boosting its artificial intelligence capabilities by open-sourcing a new foundation model that merges autonomous driving with embodied AI.

The model, named MiMo-Embodied, is described as the first open-source vision-language model of its kind to combine these two fields. Its technical details have been made available on AI developer platforms.

Xiaomi reports that the model has achieved top-tier performance, demonstrating superior abilities in areas like task planning for robotics and environmental perception for self-driving cars.

This move aligns with China's growing focus on embodied intelligence, which integrates AI into physical systems like robots. Xiaomi has been investing heavily in this area since it debuted robot prototypes in 2021 and 2022.

The project has been strengthened by the hiring of Luo Fuli, a former core researcher from AI firm DeepSeek.

This AI initiative comes as Xiaomi rapidly scales its electric vehicle (EV) production, having just built its 500,000th car. The company's recent financial performance has been strong, with a significant surge in quarterly revenue and profit.

Notably, its EV and new businesses segment, which includes these AI projects, generated a profit for the first time, with revenue nearly tripling from the previous year.

This growth occurs despite increased scrutiny following two fatal accidents involving its self-driving technology, which has prompted CEO Lei Jun to publicly reaffirm the company's commitment to safety.

Huawei Fires a Shot in the AI War, Open-Sources its Chip Secret Weapon

Huawei has launched open-source software designed to dramatically increase the efficiency of AI chips. This initiative is part of a broader push by Chinese tech firms to compete in global AI development despite restricted access to top-tier processors like those from Nvidia.

The software, named "Flex:ai," works by pooling and intelligently managing various processors, including GPUs and NPUs, to extract more performance from them.

It addresses a key challenge in AI computing: that smaller tasks often leave a single chip underutilized, while larger tasks require more power than one chip can provide. Flex:ai solves this by virtually splitting a single chip to run multiple tasks simultaneously or by combining idle processors from across a network to handle bigger jobs.

Built on the open-source platform Kubernetes, this orchestration tool reportedly boosts average processor utilization rates by 30%.

Huawei positions it as a critical step toward creating a self-sufficient computing ecosystem in China and "democratising AI" by making infrastructure more efficient.

The release highlights a growing trend in China, where companies are focusing on software optimisation to overcome hardware limitations. Huawei's Flex:ai appears similar to technology from Run:AI, an Israeli startup acquired by Nvidia earlier in 2024.

This software follows another recent Huawei tool, the Unified Cache Manager (UCM), which accelerates AI processes by optimising memory use. Together, these efforts demonstrate a strategic shift towards using sophisticated software to close the performance gap created by geopolitical constraints on advanced chips.

The Future of AI in Class? OpenAI Launches Official ChatGPT for Teachers

OpenAI is now targeting the education sector with a specialised version of its AI, called "ChatGPT for Teachers." This move comes after both OpenAI and Google have been offering free AI tools to students.

The new version, which will be free until 2027, is designed to provide a private and secure space for educators to develop lesson plans and manage student data. A key focus is on preventing AI errors from spreading misinformation to students.

Key features of ChatGPT for Teachers include:

This launch is part of the intensifying competition in the AI space, coinciding with Google's release of its new Gemini 3 model.

Google claims Gemini 3 offers superior multimodal capabilities and is being integrated across its services, including Search and the Gemini app, from day one. The company also introduced a new "agentic AI platform" named Google Antigravity.

Your opinion matters!

Hope you loved reading our piece of newsletter as much as we had fun writing it. 

Share your experience and feedback with us below ‘cause we take your critique very critically. 

Thank you for reading

-Shen & Towards AGI team