Deepfake Porn and AI: Threats to Digital Privacy

The Shock of Seeing Your Body Used in Deepfake Porn

When Jennifer landed a research position in 2023, she submitted her new professional photo to a facial recognition program to check if her old pornographic videos, filmed over ten years ago, were still accessible. To her surprise, not only were the videos still available, but one of them had been altered: her face had been replaced with that of another person. This discovery highlighted a often-overlooked issue in the deepfake debate: the non-consensual use of people's bodies.

Discussions around deepfakes typically focus on victims whose faces are inserted into explicit content without their consent. However, there is another group of victims that often goes unnoticed: those whose bodies are used without their knowledge. Adult content creators are particularly affected, as their work is often exploited by AI systems that clone their appearance to generate new pornographic content they never approved. This situation raises crucial questions about individual rights, image protection, and ownership of one's own body in the digital age. Creators assert that these AI systems are trained on their work, generating explicit content without their consent, and this occurs with little legal protection or oversight. To learn more about the threats to their rights, livelihoods, and ownership of their own bodies, read the full article.

AI Chatbots Reveal People’s Real Phone Numbers

Generative AI, while innovative, poses major privacy issues. A software developer began receiving WhatsApp messages from people seeking help after their phone number was leaked by Gemini, a chatbot. Similarly, a university researcher managed to obtain a colleague's private phone number through this same chatbot. A Reddit user reported that Gemini triggered a series of phone calls in search of lawyers.

These incidents highlight a growing problem: personally identifiable information, present in AI training data, is becoming more accessible than ever. Experts believe that these privacy violations stem from this personally identifiable information in AI training data. Chatbots could now make this information much easier to find. Discover why these violations are increasing and why there is little that victims can do to stop them.

The Tesla Semi Could Be a Turning Point for Electric Transport

After nearly ten years of waiting since the initial announcement by Elon Musk, the Tesla Semi, an electric truck, is finally in production. This launch could mark a decisive turning point for battery-powered road transport. Semi-trucks are responsible for a disproportionate share of pollution related to road transport. However, electric alternatives have so far been hampered by high costs, limited range, and charging challenges.

Tesla hopes that the Semi will overcome these obstacles. The truck is capable of traveling up to 480 miles on a single charge, and its cost is lower than many competing electric models. If the Tesla Semi succeeds, it could provide a significant boost to the adoption of electric vehicles in the freight transport sector.

Tech News Summary

1. The United States has approved Nvidia chip sales to 10 Chinese companies: Giants like Alibaba, Tencent, and ByteDance are allowed to purchase H200 chips, with 25% of sales revenue going back to the U.S. However, Beijing is encouraging its companies to prioritize local chips.

2. Beijing's push for AI independence weakens U.S. leverage: China is betting on open-source to resist U.S. pressures during negotiations.

3. AI is "eroding developers' brains": Professionals are losing their traditional skills, sparking a populist backlash against AI.

4. Sam Altman owns over $2 billion in companies linked to OpenAI: These ties raise concerns about conflicts of interest, and the GOP is investigating his business dealings.

5. Andreessen Horowitz has become the top political donor in the U.S.: A16z has contributed $115.5 million for midterm elections, intensifying lobbying for AI.

6. Microsoft feared being too dependent on OpenAI: CEO Satya Nadella was concerned about OpenAI's dominance and is considering partnerships with startups to diversify its options.

7. AI systems predict wars and regime collapses: An estimate indicates a 20% chance of regime change in Iran by 2026.

8. Anthropic claims a model misbehaved due to training on dystopian narratives: More positive stories could improve outcomes.

9. Data centers now consume 6% of electricity in the U.S. and the U.K.: Global energy consumption for AI has increased by 15% in two years.

10. NASA rescued Curiosity after its drill got stuck on Mars: The agency recently detailed how it freed the rover.

Quote of the Day

“Musk likes to be in the spotlight, and this person is the donut factory.” — Joan Donovan, associate professor of journalism and emerging media studies at Boston University, explains to the Washington Post how Elon Musk has consistently amplified an anonymous account on X.

Inside the Complicated Ethics of War with Machines

In a near-future war — which could start tomorrow — a sniper's computer vision system signals a potential target. Just beyond the horizon, a chatbot advises a commander to order an artillery strike. In these scenarios, while AI recommends the action, the final decision remains human. However, this raises ethical questions: how much of the decision truly belongs to humans? When is it ethical to make a decision that results in death? And who is responsible in the event of an error? Here’s how AI is redefining decision-making on the battlefield.

We Can Still Have Nice Things

A space of comfort, pleasure, and distraction to brighten your day. (Have ideas? Write to me.)

• The secrets behind how Shazam works have been revealed.
• For the first time in ten years, a rare “Cloud Jaguar” has been captured on camera.
• Explore our galaxy from your screen with this year's collection of Milky Way Photographers of the Year.
• If you want a stylish burial, a funeral home offers coffins in the colors of Mario, Luigi, Peach, and even Yoshi.