«

»

A new patent could position Google as the world’s dominant identity platform

By Simon Davies

Earlier this week Google was granted a US patent that could position the company as the world’s dominant identity platform with the potential to control hundreds of millions of personal identities. The implications – both beneficial and threatening – are significant.

The technology creates a link between a purported “real” identity used in social networking and one where a “persona” or “alter ego” is used.

Superficially, the concept behind the patent appears benign enough. The patented system has the ability to create different pseudonymous identities for users, meaning that users could decide who gets to see their real identity as opposed to a pseudonym, but with each identity secretly linked and thus carrying an equal degree of integrity. That means a person could establish a more flexible and trusted relationship with other users without disclosing a real world identity.

Unraveling an 11,000+ word patent document is never a simple task, but in summary it appears Google has found a means whereby a user’s identity can be assessed and then granted specific rights and permissions (though it has to be said that the words “score” and “rate” are not used in the main patent document).

The technology creates a link between a purported “real” identity used in social networking and one where a “persona” or “alter ego” is used. The system appears to use content and hehavioural analysis to recognise where a persona and a real identity are one and the same. It appears to achieve this by generating an algorithm from a variety of data feeds and processing operations.

The system appears to use content and hehavioural analysis to recognise where a persona and a real identity are one and the same.

Providing customers with the ability to use a number of accredited pseudonyms could be a powerful feature, but the system’s strategic and economic value to Google might be far greater. The uses to which a patented system can be put are of course often a world apart from the uses specified in the patent documents. In theory this system could help establish the trustworthiness of identities in the offline world and become a gateway for online services inside and outside of Google.

I’ll paste the rather arcane patent abstract here for reference:

“A system and method for generating a plurality of personas for an account of a user is disclosed. The present invention uses an account engine to receive information for the plurality of personas and to associate the information for the plurality of personas to the account. The information for each of the plurality of personas includes a visibility level. A persona engine receives a selection from the user and transmits the selected persona to the user based on the selection. An authority engine determines an appearance of the selected persona based on the visibility level.”

The patent focuses on the user advantages of the technology but of course it follows that Google would also be in a position to decide under what conditions a person can use a pseudonym and the extent to which a real identity is required.

The patent focuses on the user advantages of the technology but of course it follows that Google would also be in a position to decide under what conditions a person can use a pseudonym and the extent to which a real identity is required.

When you create or “assert” an identity (name etc) on, say, Google Plus, your behaviour and activities could be combined to assess the reputational worth and integrity of that identity.

The twist of course is that under the conditions of Google’s new amalgamated privacy policy the company would be in a position to link personas on Google Plus with personas operating on other Google services by using content and analysis from those services (would someone from Google please get back to me if I’ve misinterpreted the patent documents here).

It has to be said that not all patents see the light of day. Some become commercially obsolete during the course of approval and some are created to obstruct further innovation. At this stage it’s unclear whether Google intends deploying the technology. Nonetheless, it is important at this stage to consider the implications.

The experts who know about this are already divided on the implications. My colleague Roger Clarke in Australia believes the dangers are considerable. Google, he believes, is seeking to become the dominant manager of identities but the resulting centralisation of data “would provide massive power over individuals, to the company, to its strategic partners, and to the US Government – which would inevitably be a major user of the scheme – and to whichever governments the US regards as its allies at any given time.”
He says Google would be able to use the patent to interfere with the efforts of organisations that want to offer alternative identity management schemes that might be more freedom-friendly (this of course is an issue that goes to the heart of patents at a general level).

Other experts currently take the view that a layered approach to identity – with different levels of “real identity” assurance being established for use of different products – is a positive development, and one that is necessary to ensure trust.

It is worth bearing in mind that while this mechanism may offer some advantages to users it could also generate the most detailed body of sensitive personal information ever assembled.

In this view a more robust assessment would be justifiably conducted in environments such as social networking, while a more relaxed process would be engaged for less sensitive environments such as YouTube viewing.

My own view at this early (and quite theoretical) stage is that the two positions are not mutually exclusive. The key issues will be safeguards, transparency and accountability. Now that Google has allowed itself to treat data generated from across the company as a homogenous entity what safeguards can be put in place to ensure that the data used to assess identities does not become a “dossier”?  Will users be prohibited from creating their own “non official” identities?

It is worth bearing in mind that while this mechanism may offer some advantages to users it could also generate the most detailed body of sensitive personal information ever assembled. And that data will be available on demand to police, government authorities, courts and intelligence agencies.

I hope that if Google does decide to deploy this technology that it does so on the foundation of a wide ranging consultation rather than a mere beta test.