Waymo may use interior camera data to train generative AI models, sell ads

Waymo is preparing to use data from a robotaxis, including the video of the interior cameras associated with the identities of the contestants, to train the modular dedication models, according to an unpublished version of the privacy policy that researcher Jane Manchon Wong found.

The language project that reveals Waymo also reveals this data to customize ads, which raises new questions about the amount of contestant’s behavior within independent vehicles that can be reused to train and market artificial intelligence.

The privacy page states the following: “Waymo may share data to improve their functions and analyze them to customize products, services, ads and offers for your received.

This language is standard in today’s world; Bring cameras to the mixture is what connects the crawl factor.

Waymo gives the cyclists the option to prevent their personal information, as specified under California’s privacy laws, from participating or selling them. Passengers can also: “cancel the subscription in Waymo, or its subsidiaries, using your personal information (including the internal camera data associated with your identity) for training [generative AI]”

It is not clear what internal data can be used to train obstetric artificial intelligence models, or what are the intended cases of these models. Nor is it clear what kind of data caught by internal cameras – facial expressions? Body language? -Or whether Waymo is using data to train internal models or whether to share these data with other alphabet companies that work on artificial intelligence such as Google or DeepMind.

Techcrunch communicate to Waymo for more information and will update this post if the company responds.

Waymo, so far, is the only autonomous vehicle company that attracts Robotaxi revenues in the United States. As of February, the company records more than 200,000 Robotaki driven trips every week through its commercial services in Los Angeles, San Francisco, Vinix and Austin. This rises from 10,000 trips per week just two years ago and is a harbinger of further growth with Waymo’s expansion to new markets. The company aims to launch a commercial service in Atlanta, Miami and Washington, DC, over the next two years.

Despite these gains, Waymo is most likely losing money for Alphabet, and this may be the reason that the company explores other revenue flows, such as advertising inside the car and data sharing in artificial intelligence models.

Last year, Alphabet poured another $ 5 billion in Waymo, and the company raised an additional $ 5.6 billion of external investors who strengthened its evaluation to more than $ 45 billion.

Waymo still invests heavily in research and development and causes expansion costs, including the development of its fleet, the purchase of specialized equipment, vehicle maintenance, and infrastructure charging.

It is not clear to what extent and emo away from collapse, and less profitable. Alphabet does not explode the financial statements in Waymo in its profit report. Instead, Waymo is included in the “Other Bets” section of Alphabet in its public budget, which in 2024 recorded an operating loss of $ 1.2 billion.