Ahead of Sunday’s rumoured announcement of the HoloLens 2 headset at Mobile World Congress on Sunday, a group of concerned employees at Microsoft have begun circulating a letter demanding the august tech firm sever an augmented reality contract it won with the Army last year.
The letter—which spread across internal discussion boards and company email today—continues a trend of worker backlash within tech giants that began nearly a year ago. Googlers lobbied their bosses to kill a Pentagon drone AI contract, stop work on a censored Chinese search product, and strip forced arbitration agreements out of their employment contracts; Microsoft, Amazon, and Salesforce workers delivered similar letters urging their companies to cease any form of collaboration with Immigration and Customs Enforcement.
Microsoft secured the $480 million (£370 Million) Integrated Visual Augmentation System (IVAS) contract in November of last year, beating out other headsets like Magic Leap as well as defence sector old hands Raytheon and Lockheed Martin. Proposed uses for Microsoft’s mixed reality glasses include providing soldiers with thermal imaging, night vision, vitals monitoring, and unspecified artificial intelligence functions. Bloomberg reported at the time that up to 100,000 HoloLens devices could be supplied to the Army, but that live combat uses like IVAS were a departure from prior training application the HoloLens had seen in the military.
“The contract’s stated objective is to ‘rapidly develop, test, and manufacture a single platform that Soldiers can use to Fight, Rehearse, and Train that provides increased lethality, mobility, and situational awareness necessary to achieve overmatch against our current and future adversaries.’. Microsoft intends to apply its HoloLens augmented reality technology to this purpose,” Microsoft workers wrote. “We did not sign up to develop weapons, and we demand a say in how our work is used.”
One employee involved in drafting the letter claims it has garnered over 50 signatures. (Microsoft employs around 130,000 people.) It’s not known what number of signatories, if any, are directly involved in this project or the HoloLens more broadly.
When Google was caught working on Projects Dragonfly and Maven, workers cited the company’s own AI ethics principles which they felt should have precluded any work of that sort. Microsoft, unfortunately, has no analogous document. Instead, it uses an AI review board called AETHER (AI and Ethics in Engineering and Research), composed not of independent ombudsmen but of top executives, acting on tenets which are apparently unknown to most employees of the company. “Although a review process exists for ethics in AI,” employees wrote, “it is opaque to Microsoft workers, and clearly not robust enough to prevent weapons development.”
Where Amazon’s cloud contracts with government agencies or even Google’s aborted drone AI efforts aid military interests at an arm’s length, IVAS is closer to actual warfare than most of the big five tend to tread. This letter, which demands the contract’s cancellation, is similarly unprecedented in its full-throatedness. “As employees and shareholders,” it states, “we do not want to become war profiteers.”
We’ve reached out to Microsoft for comment, as well as insight into how its AI ethics board operates. The letter is reproduced below in full:
Dear Satya Nadella and Brad Smith,
We are a global coalition of Microsoft workers, and we refuse to create technology for warfare and oppression. We are alarmed that Microsoft is working to provide weapons technology to the U.S. Military, helping one country’s government “increase lethality” using tools we built. We did not sign up to develop weapons, and we demand a say in how our work is used.
In November, Microsoft was awarded the $479 million Integrated Visual Augmentation System (IVAS) contract with the United States Department of the Army. The contract’s stated objective is to “rapidly develop, test, and manufacture a single platform that Soldiers can use to Fight, Rehearse, and Train that provides increased lethality, mobility, and situational awareness necessary to achieve overmatch against our current and future adversaries.”. Microsoft intends to apply its HoloLens augmented reality technology to this purpose. While the company has previously licensed tech to the U.S. Military, it has never crossed the line into weapons development. With this contract, it does. The application of HoloLens within the IVAS system is designed to help people kill. It will be deployed on the battlefield, and works by turning warfare into a simulated “video game,” further distancing soldiers from the grim stakes of war and the reality of bloodshed.
Intent to harm is not an acceptable use of our technology.
We demand that Microsoft:
1) Cancel the IVAS contract;
2) Cease developing any and all weapons technologies, and draft a public-facing acceptable use policy clarifying this commitment;
3) Appoint an independent, external ethics review board with the power to enforce and publicly validate compliance with its acceptable use policy.
Although a review process exists for ethics in AI, AETHER, it is opaque to Microsoft workers, and clearly not robust enough to prevent weapons development, as the IVAS contract demonstrates. Without such a policy, Microsoft fails to inform its engineers on the intent of the software they are building. Such a policy would also enable workers and the public to hold Microsoft accountable.
Brad Smith’s suggestion that employees concerned about working on unethical projects “would be allowed to move to other work within the company” ignores the problem that workers are not properly informed of the use of their work. There are many engineers who contributed to HoloLens before this contract even existed, believing it would be used to help architects and engineers build buildings and cars, to help teach people how to perform surgery or play the piano, to push the boundaries of gaming, and to connect with the Mars Rover (RIP). These engineers have now lost their ability to make decisions about what they work on, instead finding themselves implicated as war profiteers.
Microsoft’s guidelines on accessibility and security go above and beyond because we care about our customers. We ask for the same approach to a policy on ethics and acceptable use of our technology. Making our products accessible to all audiences has required us to be proactive and unwavering about inclusion. If we don’t make the same commitment to be ethical, we won’t be. We must design against abuse and the potential to cause violence and harm.
Microsoft’s mission is to empower every person and organisation on the planet to do more. But implicit in that statement, we believe it is also Microsoft’s mission to empower every person and organisation on the planet to do good. We also need to be mindful of who we’re empowering and what we’re empowering them to do. Extending this core mission to encompass warfare and disempower Microsoft employees, is disingenuous, as “every person” also means empowering us. As employees and shareholders we do not want to become war profiteers. To that end, we believe that Microsoft must stop in its activities to empower the U.S. Army’s ability to cause harm and violence.
Featured image: Elaine Thompson (AP)