A few months ago, I came home to find my two-year-old daughter interacting with Amazon Alexa in a troubling way. Not only was she unable to distinguish it from a human, but she was also speaking in a manner that would be considered rude by any social convention. My daughter had learned to bark orders at Alexa — and that “she” (as Alexa’s voice is female) would dutifully reply no matter how rude the request. It was a jarring moment.

Could these kinds of behaviors toward our in-home voice assistants seep into other areas of our lives? Dr. Sheryl Brahnam, assistant professor of computer information systems at Missouri State University, thinks so. She posed a stark warning to us all when she wrote, “If our behavior towards AI is mimicked by our children and influences our treatment of each other, there could be significant implications for gender dynamics.”

AI is unlike any other technology we have created before, because it’s personified with a strongly implied gender. And the gender bias that’s permeated history has already reared its ugly head. Every major home assistant — Apple’s Siri, Amazon’s Alexa, Google Home, and Microsoft Cortana — is given a female voice and docile, obedient personality. The AI revolution has, until now, been shaped by men and is starting to marginalize half of the world’s most critical stakeholders: women. Shame on us.

Or, I should say: shame on us again. Creating this type of male-led workplace and product bias is not a new trend for the tech industry. Most of the world’s major technology companies are led by male CEOs, as they have been for the last 30 years of the tech boom, and over 85% of their engineers are also male. Margaret Mitchell, while a researcher at Microsoft, coined a term for these kinds of workforces — a “sea of dudes” — and warned they were “putting themselves in a position of myopia.” Yet we continue to make the same mistakes over and over.

Having founded my company in 1995, I lived through the heyday of launching websites, search, and then social, and I remember how the industry rushed forward in a series of land grabs, at a fever pace, without thinking through the potential misuses of these technologies or the harm they could cause us, as individuals and communities. We’re now seeing where that lack of forethought gets you, and the results are disruptive and disturbing. Just look at 2016’s presidential election. It’s a mess.

In the technology industry, these feverish land grabs have been driven by mostly male “echo chambers” — a term now heavily, maybe inextricably, linked with Silicon Valley. They are ubiquitous and glaring. The largest conference for AI is undertaken by NIPS, which recently reported that only 17% of its attendees were women in 2017. This is not only wrong and unfair, but it’s also bad for business.

There is a real risk today that some of the most important decisions governing the future of humanity will be made by a small group of men who come from similar backgrounds and bring similar perspectives. How can we make huge decisions that have global implications if women — half of the world’s population — are missing from the equation? The results could be catastrophic.

We know we need to employ more women across industry, but we — myself included — are still fumbling. I work judiciously with my team to make diversity one of our top priorities at LivePerson, but it still isn’t where we want it to be.

To address the challenge of AI, we need a body of diverse leaders, led by women, to come together and create industry-wide standards for AI, virtual assistants, and chatbots. The Turing test, developed by Alan Turing in 1950, posited how a machine could mimic the intelligence of a human. That’s no longer good enough by itself. As bots come closer to passing this test, they can mimic negative human traits as well — such as hidden biases or a lack of empathy. The positive potential of AI is endless and exciting, but until the tech industry sets standards and begins to hold itself accountable, we’ll continue to miss the diversity of perspective and inclusivity we need for true technological progress.

Perhaps the most amazing scientist of our time, the late Stephen Hawking, said it best: “Success in creating AI would be the biggest event in human history, but it could also be the last, unless we employ best practice and effective management.” Just as we are entering what many would describe as a watershed moment in addressing multi-sector gender bias, the tech industry is in a danger zone.