Microsoft wants regulation to keep the 'facial recognition genie' in the bottle

There's no time like the present to slow the tide of a technology-enabled facilitation of discrimination aided by facial recognition.

So, somewhat surprisingly, argues one key developer of said facial recognition tech: Microsoft.

On Dec. 6, Microsoft president Brad Smith released an over 3,500-word blog post detailing his concerns with the aforementioned technology and calling for immediate government regulation. The post gets straight to the point, noting that as a society we can't risk waiting any longer.

"The facial recognition genie, so to speak, is just emerging from the bottle," reads the post in part. "Unless we act, we risk waking up five years from now to find that facial recognition services have spread in ways that exacerbate societal issues."

When the president of what is, or at least for a moment in late November was, the world's most valuable company publicly decries the potential future harms of a tech we should probably take note.

This rings true especially when that very same company has a $480 million contract with the U.S. military to help it "increase lethality." Because if facial recognition tech spooks even Smith, then the rest of us should be feeling pretty nervous right about now.

Specifically, the blog post lays out three problems with facial recognition tech.

First, it "[increases] the risk of decisions and, more generally, outcomes that are biased and, in some cases, in violation of laws prohibiting discrimination." This is a very real concern, as we know that in its current form the technology has a hard time correctly identifying people of color, women, and young people.

Next, "the widespread use of this technology can lead to new intrusions into people’s privacy." We've seen what this looks like in China, with automated facial recognition tech being used to identify and publicly shame jaywalkers.

And, lastly, "the use of facial recognition technology by a government for mass surveillance can encroach on democratic freedoms." This we know to be true, as the tech can identify individual protesters at, say, a march. Even, potentially, if they're masked.

This is not the first time Microsoft has asked the government to regulate the sector. The company issued a similar call in June. However, this time around, Microsoft is also asking its competitors to not wait for the government to act. "We and other tech companies need to start creating safeguards to address facial recognition technology," Smith writes.

Notably, Smith doesn't think the technology is all bad. He writes that it has been used to find missing children, and even to help diagnose a rare disease. Still, according to Smith, it clearly needs some serious and immediate checks.

Perhaps Smith, in addition to this lengthy call for regulation, can do us all one more favor and send his blog post directly to Amazon's Jeff Bezos. Maybe Bezos, whose company has its own controversial facial recognition program, will take the concerns of a fellow rich man more seriously than those of the organizations that have spoken up against it.

But no point in holding your breath — it's unlikely the blue hue of your oxygen-starved face will pose any problems to the algorithm identifying you in the crowd.