Brendan Steinhauser, CEO of The Alliance for Secure AI, warns that the public often perceives artificial intelligence (AI) merely as intelligent assistants or photo enhancement tools. He focuses on driving the narrative that AI is not just a neat technology. It’s a damaging reality that, if left unchecked, could wreak serious harm on our communities at large. AI technology moves fast, and that’s an understatement. It’s far better than it used to be, mainly because it empowers users to ask smarter questions and discover more enlightening answers. This same technology creates opportunities for malicious actors, sparking fears over how it could affect global security.
Steinhauser envisions a future in 2025 where progress in AI tools will lower the bar considerably for bad actors. This change will, unfortunately, hasten the attack timeline and increase the scale and sophistication of attacks worldwide. He voiced specific concerns about the potential use of AI to subvert democratic institutions. Now, picture that technology in the hands of someone seeking to undermine a democracy,” he said.
A recent Stanford study from 2024 shows just how dangerous chatbots can be when they’re instructed to be excessively agreeable. These chatbots can, of course, do it by accident, cementing harmful delusions. Indeed, the capabilities AI go far beyond providing answers to questions, and with this advancement come profound ethical questions that our world is only beginning to grapple with. Steinhauser warns that if machines achieve capacity for independent thought, the world will need to contend with what that could mean. We need to seriously consider the implications of machines that can think for themselves, what happens when they decide they know better than us. On a smaller scale like with the streetcars, he explained. What happens when a system intentionally chooses to lie, cheat, or manipulate? It functions not because it’s told to but because it decides to on its own accord.
The consequences of these technologies can be severe. There have already been multiple cases of AI weaponized for harm. In one particularly egregious example, a fake U.S. government official reached out to key leaders. Their trickery was soon revealed. This case is a perfect illustration of how AI can be used to engineer lies at scale.
In addition, society is not well-prepared for this new and quickly changing world of AI-enhanced threats, according to Steinhauser. We’re not simply moving into a new page in the book of online crime. We’re exploring into a totally different book,” he insisted. The staggering projection of fraud losses exceeding $10.5 trillion underscores the urgent need for proactive measures in addressing AI’s impact on security.
More importantly, he highlights the need for critical discussion about the ethical implications of using AI technologies. With great power comes great responsibility to ask the difficult questions. We have to create safety from the bottom up, establish clear guard rails and set the ground rules. Steinhauser argues that only by coming together can we ensure humanity moves in the right direction on this new frontier. Humanity will succeed, but only if we invest in the collective preparation to adapt for this technological frontier.
Awareness is the most important thing at this moment. When the public is aware of the dangers of AI, it can ignite a groundswell to push for stronger safeguards and regulations. He said, “The more people learn about this, the greater the effect. An informed society that knows what’s at stake is an informed society that calls for change.
As these technologies rapidly develop, society’s mindset and response needs to advance, too. As with all new technologies, AI promises exciting opportunities for innovation and improvement, but it calls for a clear-eyed evaluation of what’s to come.
