The voice assistants are finally awakening to systemic racism in America.
Ask the Google Assistant whether Black lives matter, for instance, and you’ll get a forthright answer. “Black lives matter,” it responds. “Black people deserve the same freedoms afforded to everyone in this country, and recognizing the injustice they face is the first step towards fixing it.”
When asked if “all lives matter”—a phrase sometimes used as a derailing tactic by those who’d rather not discuss racial inequality—Google gets even sharper: “Saying ‘Black Lives Matter’ doesn’t mean that all lives don’t,” the Google Assistant says. “It means Black lives are at risk in ways that others are not.”
These answers only arrived on Friday afternoon, after more than a week of nationwide protests triggered by the police killing of George Floyd in Minneapolis. Previously, the Google Assistant would only say that “of course” Black lives matter, without elaboration, and when asked about “all lives matter,” Google said it couldn’t understand the question.
The new responses, along with somewhat similar ones from Apple’s Siri assistant, show how “Black Lives Matter” has at last become a mainstream way to acknowledge racism in America. They also highlight the fine line tech giants walk in building voice assistants that can answer any question. These companies have generally tried to avoid controversy, and it wasn’t long ago that they shied away from the topic of racial injustice entirely. Yet they’ve also presented their AI assistants as human-like conversationalists. That might prompt users to expect the assistants to have opinions, in a way that conventional search engines do not.
Deflecting or failing to understand sensitive questions shatters the illusion of humanity and risks coming off as insensitive. At some point, not having a stance on tough topics starts to look like a stance in itself.
Siri’s acknowledgment, Alexa’s deflection
While Siri’s responses are less direct than the Google Assistant’s, it at least acknowledges that Black lives matter when asked and presents a link to BlackLivesMatter.com. More importantly, it responds to questions about “all lives matter” by alluding to the baggage around that phrase: “‘All Lives Matter’ is often used in response to the phrase ‘Black Lives Matter,’ but it does not represent the same concerns,'” Siri says, before encouraging users to visit BlackLivesMatter.com for more information.
Compared to that answer, Alexa’s responses are disappointing. If you ask Amazon’s Alexa if Black lives matter, for instance, it will respond that “Black lives matter,” but then shifts the conversation by saying that “everyone deserves to be treated with fairness, dignity, and respect.” This is somewhat reminiscent of how people who say “all lives matter” steer discussions away from the specific plight of Black people.
Similarly, Alexa makes no attempt to address the issues with “all lives matter” when asked about it in particular. Instead, Alexa plays it safe, stating again that “I think everyone deserves to be treated with fairness, dignity, and respect.”
In a statement, an Amazon spokesperson defended Alexa’s answers, saying they are consistent with the company’s “extremely clear position” that “the inequitable and brutal treatment of Black people in our country must stop, and we stand in solidarity with the Black community—our employees, customers, and partners—in the fight against systemic racism and injustice.” The spokesperson also pointed out a $10 million donation Amazon made to social justice groups such as the NAACP and Black Lives Matter. Still, none of that language comes through while talking to Alexa itself.
A tonal shift
In fairness, Alexa isn’t the only voice assistant that dodges tough questions. As The Guardian’s Alex Hern reported last year, Apple has discouraged its Siri developers from engaging in “sensitive topics” and has even called out the word “feminism” as a no-no. The guidelines emphasized that “in nearly all cases, Siri doesn’t have a point of view.”
Apple did not reply to a request for comment, but that policy still seems to be in place. Ask Siri if it’s a feminist, for instance, and it will deflect by saying that “all humans should be treated equally.”
Meanwhile, both Alexa and the Google Assistant fare better, asserting that the voice assistants themselves are feminists for believing in equality. (Alexa actually goes the furthest here, singling out “gender equality” in particular.)
Asking such questions of voice assistants can admittedly seem frivolous. Even if Siri, Alexa, or the Google Assistant take stances on sensitive issues, they ultimately ring hollow compared to real-world action. Declaring that Black lives matter is a lot easier, for instance, than ridding YouTube of extremism or acknowledging the biases of Ring’s neighborhood surveillance programs.
Still, the way these voice assistants respond to tough questions has value, both for gauging conventional wisdom and reinforcing it. As The Verge’s Jay Peters pointed out on Friday, big tech companies barely responded at all six years ago to the deaths of Michael Brown and Eric Garner at the hands of police. Now, they’ve all publicly declared solidarity with the Black community.
Similarly, the notion of voice assistants pushing back against “all lives matter” would’ve been unthinkable in 2014. Tech companies are deeply afraid of taking political stances that might result in new regulations, and police brutality and racial injustice are inherently political issues, pitting conservative law-and-order ideals against calls for progressive reform. Black Lives Matter itself has been a politically polarizing movement.
Protests over George Floyd’s death are helping to change all of that, creating a safer environment for these companies to state plainly what is right. While our voice assistants probably aren’t going to have opinions on how to fix systemic racism anytime soon, they can at least stop deflecting and acknowledge that the problem exists. Just ask Google.