In this interview, AI godfather Yoshua Bengio discusses attention-grabbing headlines about AI, taboos among AI researchers, and why top AI researchers may disagree about the risks AI may pose to humanity.

In order to achieve the ultimate goal of a world without nuclear weapons, it may be necessary to first achieve another goal—eliminating the reasons why nuclear weapons were developed in the first place.

From deciding which soldiers to enhance, to figuring out which enemies to disable, tomorrow’s military leaders will face a host of thorny moral dilemmas.

Current modernization plans for the US strategic nuclear triad may not have sufficiently considered the range of available options made possible by new technology, or adequately considered whether a more streamlined and less expensive force structure would suffice for deterrence.

The incentives and practicalities make it unlikely that robotic car bombs will be a near-term threat.

The nuclear weapons production and laboratory system created during the Cold War is simply far too large for the current military situation and needs drastic consolidation that includes the closing of labs and other facilities

The longtime New York Times national security reporter talks about his new history of cyberwar, why we need a public debate, and how cyberattacks make nuclear war more likely.

As the lame-duck Congress wraps up business, a serious debate is unfolding over the future of the US nuclear weapons complex. For the first time since the end of World War II, the long-held policy that places control of the design and production of nuclear weapons in civilian hands may be up for grabs. At issue: What is to be done with the National Nuclear Security Administration (NNSA), now located inside the US Department of Energy?

During the 1970s and 1980s, the size of the U.S. nuclear weapons budget allowed for a healthy amount of high-risk, long-term basic research to be conducted at the nation's nuclear weapons laboratories. Much of this research grew out of, but diverged from, the labs' core weapons-related capabilities. Importantly, the wide-ranging capabilities at these laboratories allowed other national security agencies to tap into that expertise on an "as needed" basis without making the long-term investments necessary to build and sustain it.

If the United States decides to send cluster munitions to Ukraine, it should consider investing in autonomous capabilities for demining.

The Defense Nuclear Facilities Safety Board faces difficulties that include the actions of some if its own members, who either don’t want or can’t seem to execute its mission. Last year, Sean Sullivan, the acting chairman, tried to secretly convince the Trump White House to get rid of the board entirely, claiming it was “a relic of the Cold-War era defense-establishment.” Sullivan failed and was compelled to resign, but the board has adopted a backup plan that imposes large budget and staffing cuts on the safety agency.

Countries around the world are modernizing their nuclear weapons systems. Will they insert more powerful AI into the controls?

Once upon a time not so long ago, glossy posters, press releases, and other public relations materials flooded congressional offices and newsrooms inside Washington's Beltway, touting the futuristic antimissile project known as the Airborne Laser (ABL). The publicity campaign included an artist's rendition of a Boeing 747, modified so a reddish laser beam shot from its nose turret. As if they were publicizing a blockbuster movie, ABL posters carried the words "Coming soon to a theater near you."

Increasing automation within nuclear weapon command systems means putting faith, and lives, in the hands of algorithms that may never fully understand.

The cost of the nuclear weapons complex keeps going up, even as the size of the nuclear arsenal falls

A wave of declassification is shedding light on French nuclear weapons development, especially in Polynesia, although some gaps remain.

If the Obama administration does not put in place an affordable nuclear weapons strategy for the coming decades, nuclear strategy will be set by bureaucratic struggles and congressional politics. This is not strategy; it is an accident waiting to happen.

The recent India-China border clashes demonstrate the ugliness of the “stability” that nuclear weapons have brought to Southern Asia.