Skip to content
Search

Latest Stories

Follow Us:
Top Stories

The U.S. election system was already wobbling, and now here comes AI

people voting

In some states, more than half of election officials have quit, writes Klug.

Brett Deering/Getty Images

Klug served in the House of Representatives from 1991 to 1999. He hosts the political podcast “ Lost in the Middle: America’s Political Orphans.”

As we head into election season, the potential for misinformation is enormous and the ability of election officials to respond to artificial intelligence is limited.

The new technology arrives at a time when we still haven’t gotten our arms around social media threats.


“The ability to react at the pace that's being developed is almost impossible,” worries Idaho Secretary of State Phil McGrane, a Republican. “By design, our system is meant to be slow and methodical.”

While deep fakes get all the attention, the truth is the threat arrives at a time when election administration itself is shaky. Election officials caught the brunt of the mistrust. In some states more than half of them have quit.

“I have a little PTSD, as do my coworkers,” said Nick Lima, who heads up elections in Cranston, R.I., and who — with some reservations — decided to keep the job he loves. “During election season, you know, you really feel the pressure, you feel your heartbeat increasing a bit.”

Today everyone who works in the campaign infrastructure faces unending scrutiny. If you thought it was easier in a red state, you are mistaken.

“Just the act of standing behind you watching you work just puts you on edge,” said McGrane. “Now you start second-guessing yourself, even if you know you're doing it right. The poll workers don't know about cyber security on voting equipment, but your poll watcher is getting asked these questions. “

Deep fakes are one level of concern, but Edward Perez, who had been director of civic integrity at Twitter and is now a board member at the OSET Institute (whose mission is to re-build public confidence in our voting system), worries about the misuse of AI to disrupt the backroom of every American precinct.

“One of the most important things to understand about election administration is, it’s very, very process oriented. And there’s a tremendous number of layers,” he said. “Are we talking about voter registration? About the security of election administration? All of this technology is never deployed just in a vacuum.”

The fact that the election system is a conglomeration of different rules and regulations from 50 different states with 50 different voting rules adds to the complexity. The challenge is serious as election officials scramble in this election-denying climate to staff 132,000 polling places with 775,000 volunteers. The clock is ticking to deploy the necessary defenses against threats that aren’t fully understood.

From hanging chads to deep fake videos, American democracy wobbles by Scott Klug

Read on Substack

Read More

This 3D rendered image shows a central AI processing chip sitting atop a glowing blue printed circuit board.

Can AI profit-sharing help workers? Examining public wealth funds, AI taxes, economic transition policies, and the future of work.

Jason marz / Getty Images

There’s No Easy Path Through the AI Transition

“Trending” policy ideas tend to garner attention for all the wrong reasons: they seem like silver bullet solutions that will save us from taking on much harder reforms. Proposals to share profits from leading AI companies with the public are the latest example. It’s the rare policy scheme that seems to have united President Donald Trump, Senator Bernie Sanders, and CEOs at the leading AI labs. While the proposals for AI profit sharing vary in their precise details, a quick review of their likely outcomes should quickly deflate the popular excitement that has formed in response to calls for new taxes, public wealth funds, and the like. It’s important to reveal such limitations so that the AI policy discourse can move on to mechanisms more likely to address the real concerns of the American people.

In our first hypothetical world, the two leading AI labs—Anthropic and OpenAI—give away 3% of their equity. That’s not nothing! Based on current figures, such a contribution could kickstart a public wealth fund of about $55 billion. Let’s then imagine that fund earns 10% a year (a big “if” but let’s run with it). Per The Economist, this AI would reach a staggering $140 billion within ten years. How much would that benefit Americans? If annual payouts were 4% — what the publication reports is a proper amount to keep the fund going and growing — Americans would have an extra $20 in their pocket.

Keep ReadingShow less
For Imre Huss, Fixing Democracy Starts With Talking to a Stranger
a couple of people sitting at a table with cups of coffee

For Imre Huss, Fixing Democracy Starts With Talking to a Stranger

The Democracy Architects Council, presented by The Bridge Alliance Education Fund and Civics Unplugged, offers a paid, one-year fellowship for eight fellows ages 18 to 28, each selected for their work across a distinct sector of democratic life.

The youngest member of the Democracy Architects Council is building AI-powered civic tech, but he says the real work of democracy still happens face to face.

Keep ReadingShow less
Vote Badge with Rising Social Media Like Icons and Hearts – Digital Engagement and Online Voting
J Studios / Getty Images

Democratic Autopsy and AI

After every defeat, organizations conduct autopsies. The good ones are honest, like NASA’s Rogers Commission report after the Space Shuttle Challenger exploded shortly after takeoff. In addition to identifying the infamous O-rings as the proximal culprit, it looked at organizational culture, communication failures, normalization of risk, management pressures, and institutional blind spots. The best ones are uncomfortable, and make a serious effort to understand “why did we mess this up so badly?” I’ve personally seen both good “autopsies” and bad ones throughout my decades of experience in true life-or-death realms: the SEAL Teams and as an Emergency Medicine physician.

Following the 2024 election, the Democratic National Committee produced a lengthy report titled Build to Win. Build to Last. Yet it is not a serious document because it does nothing to prepare for the unstoppable and very near future staring us right in the face. It is nearly 200 pages long and attempts to explain what went wrong and how the party should prepare for the future. It discusses organizing, communications, coalition building, fundraising, digital strategy, and voter outreach. It is filled with references to data, analytics, and technology.

Keep ReadingShow less
My Generation Can Spot the Deepfake. That’s Not Enough.
Smartphone with ai text in jeans pocket
Photo by Immo Wegmann on Unsplash

My Generation Can Spot the Deepfake. That’s Not Enough.

Thomas Massie, a seven-term Republican congressman from Kentucky, lost his primary on May 19. The race cost $32.6 million, making it the most expensive congressional primary in U.S. history. Among the weapons deployed against him: an AI-generated video showing him checking into a hotel room with Representatives Alexandria Ocasio-Cortez and Ilhan Omar, with their hands clasped. The narrator called it "worse than adultery." A disclaimer at the bottom of the screen, in small text, read: "This satirical ad was created with artificial intelligence."

I watched the ad. It looks ridiculous. The movements are slightly too smooth, the lighting is off, and the scenario is so cartoonish that I genuinely could not tell at first whether it was meant to be taken seriously. But I'm 17, and I've spent the last four years watching AI-generated content get better in real time. I know what the seams look like. Massie, in his post-loss interview on Meet the Press, was blunt about who the ad actually reached: "It was actually very effective on the boomers."

Keep ReadingShow less