General Motors has begun advertising the integration of Siri in the Chevrolet Sonic. The ad is a work of art (http://www.youtube.com/watch?v=g4aVDp-OQaA) on a par with the most clever advertising output from Apple, which is a good thing for GM which is hoping some of Apple’s cachet rubs off on the Sonic.

Siri does not have a great history of excellent performance when used in a car, which raises questions regarding the efficacy of this integration and the ROI for GM.  If Siri doesn't boost sales or customer satisfaction scores, this may be a brief experiment. 

One has to salute GM for its willingness to take a risk here.  If Apple showed even a hint of being a more reliable partner, this analyst would be more enthusiastic, but it is painfully clear that automotive industry priorities are anything but a priority to Apple.

First, Siri is a hold-the-phone-to-your-mouth speech recognition technology and is not optimized for use in cars generally or in the Sonic in particular. Automotive grade speech recognition engines are tuned to individual cars and their particular aural characteristics – and most, though not all, take advantage of echo cancellation and noise reduction to enhance speech recognition.

The best speech recognition systems – a space dominated by Nuance – take into account automotive specific use cases to improve recognition by limiting or better organizing the recognition process. Systems such as Dragon Drive from Nuance or even less sophisticated versions make assumptions about what kind of questions or information needs are most likely to arise in the car to improve the chances and speed of recognition thereby mitigating distraction.

Even speech recognition systems from Nuance don’t get it right every time, but the chances of Siri mitigating distraction by enhancing speech recognition in a noisy vehicle cabin is slender indeed.  Integrating Siri is an invitation to confusion, dissatisfaction and disappointment – and that only accounts for the iPhone owners.  Owners of mobile phones from other suppliers are completely out of luck.

Which brings me to the second dimension of failure inherent in any Apple integration.  The integration of Siri in cars falls into the overflowing bucket of gimmicky automotive solutions intended to borrow some sales and marketing pizzazz from the world of technology.  Similar short-sighted strategems have included:

Google Search – BMW and others

Google Earth – Audi

Google Maps – Tesla

Apple docking cradle – BMW

Apple iPhone dock – VW

Every one of these solutions is a failure at some level for not taking into account the special circumstances of in-vehicle integration.  Google Search works – awkwardly – in appropriately equipped BMWs, but does not take into account the driving direction or route of the car – hence some results may not be in the direction of travel.  There is a clear lack of contextual awareness.

Google Earth is an impressive and expensive implementation in AudiConnect equipped cars in the U.S., but this solution seems to be more of a proof of concept than a real value proposition to Audi drivers.  Google Earth is a show-offy feature that adds nothing to the safe or efficient use of the car – and requires a $15 monthly subscription besides.

Google Maps in Tesla’s Model S is another impressive and no-doubt expensive implementation.  The live map is delivered to the 17” display in the car via a 3G embedded modem.  Tesla has only recently acknowledged its intention to provide some caching capability to overcome the tendency of the map image to pixilate or disappear entirely.

The Apple docking cradle for selected BMWs was no doubt another expensive engineering and design proposition.  In the end, the cradle – costing $250 – was only suitable for use with the iPhone 4 and only enabled access to audio or video files (projected into the head unit) purchased from the iTunes store.

BMW’s iPhone dock, in other words, allowed Apple to open an iTunes kiosk integrated in the center console.  No matter, any iPhone docked in the BMW cradle had a tendency to overheat undermining the cleverness and expense that went into creating the device.

And, finally, there's the iPhone dock for the so-called iBeetle.  According to a Mashable report the iBeetle models – in Applefied color schemes – will launch in the beginning of 2014 following their debut at the Shanghai auto show.

Writes Mashable:  “When the iPhone is connected to the docking station or synced wirelessly to the dashboard, it unlocks various features via the iBeetle app.  For example, it can read Facebook and SMS messages out loud, and, in addition to syncing with iTunes, you can switch back and forth with Spotify.

“Another features is Postcard, which sends the current location of the iBeetle to friends as a digital message. Besides the car's location, the postcard also displays engine temperature gauges, a chronometer and a compass.”

As with Apple’s announcement of its iOS in Car docking station, nowhere is there a mention of enhanced navigation, traffic data or safety – to say nothing of distraction mitigation.

Bottom line, companies such as Google and Apple do not have automotive priorities at heart or in their plans.  To put it more clearly, Apple and Google have no idea what is required to support an auto maker shipping cars around the world.  Google wants to sell advertising and search results and Apple wants to sell phones, tablets, etc.  Those are the buttons they want to push and it pushes my buttons every time.  Does it push yours too?