AI Flash News

A new training method improves the reliability of AI confidence estimates without sacrificing performance, addressing a root cause of hallucination in reasoning models.

The associate professors of EECS and chemistry, respectively, are honored for exceptional contributions to teaching, research, and service at MIT.

Founded by Tristan Bepler PhD ’20 and former MIT professor Tim Lu PhD ’07, OpenProtein.AI offers researchers open-source models and other tools for protein engineering.

Researchers are developing hardware and algorithms to improve collaboration between divers and autonomous underwater vehicles engaged in maritime missions.

As the School of Humanities, Arts, and Social Sciences marks 75 years, Dean Agustín Rayo reflects on how AI is reshaping higher education and why SHASS disciplines continue to be central to MIT’s mission.

AI-powered personas are becoming so realistic that they can infiltrate online communities and subtly steer public opinion. Unlike traditional bots, they adapt, coordinate, and refine their messaging at a massive scale, creating a false sense of consensus. Early warning signs—like deepfakes and fake news networks—have already appeared in global elections. Researchers warn that the next election could be the true test of this technology’s power.

Engineers at Northwestern University have taken a striking leap toward merging machines with the human brain by printing artificial neurons that can actually communicate with real ones. These flexible, low-cost devices generate lifelike electrical signals capable of activating living brain cells, a breakthrough demonstrated in mouse brain tissue.

Calling AI things like “smart” or saying it “knows” something might sound harmless, but it can quietly mislead people about what AI actually does. A new study shows that news writers are more careful than expected, rarely using strongly human-like language. When they do, it often falls on a spectrum—sometimes describing simple requirements, other times hinting at human traits.

Researchers have shown that blending quantum computing with AI can dramatically improve predictions of complex, chaotic systems. By letting a quantum computer identify hidden patterns in data, the AI becomes more accurate and stable over time. The method outperformed standard models while using far less memory. This could have big implications for fields like climate science, energy, and medicine.