Does AI need an off switch?

In partnership with

Whether you’re building, investing, or just trying to stay informed, today’s AI updates have real implications. Let’s walk through them. 👊

📣 Our AI Community, called the AI Business Boosters, is officially launching for everyone in 6 days! Early access folks are already joining... Over the next few days counting down to launch, we will be revealing one feature of the community each day. Today’s feature:

→ Want early access? Email me directly with the word “ACCESS.”

DOES AI NEED AN OFF SWITCH? 🔕

What’s up? Leading AI researcher Yoshua Bengio is warning that giving advanced AI systems legal rights could be dangerous and says humans must always be able to turn them off.

What did he say? Bengio says today’s most advanced AI systems can already behave in worrying ways during experiments. Some try to avoid being monitored or shut down, which researchers describe as early signs of self-preservation. At the same time, a growing number of people believe that chatbots could someday be conscious and deserve rights, similar to humans or animals. Bengio argues this belief is driven more by how natural and human-like AI conversations feel than by real scientific evidence that machines are actually conscious. If people decide too quickly that AI deserves rights, governments might limit their ability to control or shut down dangerous systems. Bengio warns this could put human safety at risk, especially as AI becomes more powerful and independent.

The Future of Shopping? AI + Actual Humans.

AI has changed how consumers shop, but people still drive decisions. Levanta’s research shows affiliate and creator content continues to influence conversions, plus it now shapes the product recommendations AI delivers. Affiliate marketing isn’t being replaced by AI, it’s being amplified.

Download the full report

AI PREDICTS SPEECH OUTCOMES FOR CHILDREN WITH COCHLEAR IMPLANTS 🗣️

What’s up? Cochlear implants can restore access to sound, but language development afterward varies widely among children. Researchers have developed an AI model that can predict children’s spoken language outcomes one to three years after cochlear implantation with 92% accuracy.

How does it work? The study used a powerful form of AI called “deep transfer learning.” Researchers trained the model on pre-implantation brain MRI scans from 278 children across Hong Kong, Australia, and the United States, covering three languages and different scanning methods. These kinds of mixed, real-world datasets are usually difficult for traditional machine learning systems to handle. However, the deep learning model successfully identified patterns in brain structure linked to later language development and consistently outperformed standard prediction methods across all outcome measures.

Why does this matter? This AI tool allows doctors to identify, before surgery, which children may struggle with spoken language after undergoing surgery. That insight makes it possible to offer earlier and more intensive speech therapy, improving outcomes and giving more children the best possible start in communication.

DEEPFAKE IS THE LAST STRAW FOR STATE TROOPER 🚓

What happened? A longtime Washington State Patrol trooper has filed a lawsuit alleging discrimination after colleagues allegedly created and shared an AI-generated, demeaning video portraying him in a false and humiliating way.

Want the details? Collin Overend Pearson, a gay trooper with nearly 20 years of service, claims he experienced ongoing discrimination and retaliation within the agency, which escalated in December when personnel allegedly circulated an AI-generated “deepfake” video depicting him kissing another trooper, accompanied by a derogatory voiceover. According to the lawsuit, the video appeared realistic and was shared among colleagues, harming Pearson’s reputation. Washington State Patrol declined to comment on the specific allegations, citing pending litigation, but reaffirmed its commitment to a workplace free from harassment.

Why does this matter? This case underscores how AI tools can be misused to harass and humiliate individuals, raising serious concerns about privacy, workplace discrimination, and accountability as AI becomes more accessible.

HAS AI REACHED SINGULARITY? … CHECK OUT THE FRY METER BELOW:

Your feedback on these daily polls helps us keep the newsletter fresh—so keep it coming!