China’s Dealer Meltdown, Stellantis Parking Police, AI Can't Actually Learn Podcast By  cover art

China’s Dealer Meltdown, Stellantis Parking Police, AI Can't Actually Learn

China’s Dealer Meltdown, Stellantis Parking Police, AI Can't Actually Learn

Listen for free

View show details

Shoot us a Text.

Episode #1303: China’s price war is crushing dealers while a quirky Stellantis policy highlights a very different retail reality here at home. Plus, a new AI test reshapes how we define intelligence


  • China’s brutal EV price war is crushing dealership profitability, with more than half now losing money. As automakers slash prices to compete, retailers are stuck selling cars at a loss just to keep up.
    • 56% of Chinese dealerships were unprofitable in 2025, up sharply from 42% the year before.
    • A staggering 82% of dealers sold new cars below cost, pushing margins to a negative 26%.
    • Financing and insurance profits also dropped after tighter lending regulations hit dealer income streams.
    • The only bright spot: service and parts, with margins soaring to 81% as dealers pivot to survival mode.
    • Outlook remains bleak, with just 23% of dealers expecting market growth in 2026.


  • At Stellantis HQ, what you drive to work might determine where you park and whether you get a warning. The automaker is reinforcing brand loyalty with preferred parking… and some awkward consequences.
    • Employees have reported getting tickets for parking non-Stellantis vehicles in preferred spots.
    • Prime parking is reserved for company brands, with violators risking warnings, or even getting booted.
    • The policy reflects a long-standing Detroit culture of encouraging employees to drive what they build.
    • Confusion happens, one employee got ticketed for parking their Eagle Talon in the right spot
    • As one observer put it, there’s “strong motivation” to drive company cars, especially when the walk can be up to 30 minutes from the farthest lots


  • A new AI benchmark just dropped: and it’s exposing a major gap between human intuition and machine intelligence. ARC-AGI-3 tests whether AI can learn on the fly. Spoiler: it can’t… at least not yet.
    • Every major AI model scored under 1%, while humans solved everything on the first try without instructions.
    • The test measures real adaptability—throwing AI into brand-new environments with zero training or prompts.
    • Critics say the scoring system is stacked, but the bigger debate is shifting to how we measure intelligence at all.
    • ARC’s creator argues current AI only works because humans build complex “scaffolding” around it. True AGI shouldn’t need that.


Today’s show is brought to you by HeyGreenlight. HeyGreenlight’s Wingman gives your sales and BDC team live, real-time guidance so they consistently say the right things, at the right time, on every call.

Join Paul J Daly and Kyle Mountsier every morning for the Automotive State of the Union podcast as they connect the dots across car dealerships, retail trends, emerging tech like AI, and cultural shifts—bringing clarity, speed, and people-first insight to automotive leaders navigating a rapidly changing industry.

Get the Daily Push Back email at https://www.asotu.com/

JOIN the conversation on LinkedIn at: https://www.linkedin.com/company/asotu/

No reviews yet