Tech
We summarize the week's scientific breakthroughs every Thursday.
- Psychology
Chatbots spewing facts, and falsehoods, can sway voters
Chatbots that dole out fact-laden arguments can sway voters. Those facts don’t have to be true.
- Artificial Intelligence
Chatbots may make learning feel easy — but it’s superficial
People who use search engines develop deeper knowledge and are more invested in what they learn than those relying on AI chatbots, a study reports.
- Animals
AI eavesdropped on whale chatter. It may have helped find something new
Some “clicks” made by sperm whales may actually be “clacks,” but marine biologists debate what, if anything, that means.
- Artificial Intelligence
As teens in crisis turn to AI chatbots, simulated chats highlight risks
From blaming the victim to replying "I have no interest in your life" to suicidal thoughts, AI chatbots can respond unethically when used for therapy.
- Chemistry
A new AI technique may aid violent crime forensics
An AI tool trained on chemical signatures from corpse-eating insects may help determine time and place of death for victims of violent crimes.
- Artificial Intelligence
The AI model OpenFold3 takes a crucial step in making protein predictions
The open-source AI model improves transparency in predicting how proteins interact with other molecules, which could speed up drug discovery.
- Artificial Intelligence
A conference just tested AI agents’ ability to do science
AI promises to speed up scientific analysis and writing. However, AI agents struggled with accuracy and judgment.
- Artificial Intelligence
Biased online images train AI bots to see women as younger, less experienced
Age and gender bias in online images feeds into AI tools, revealing stereotypes shaping digital systems and hiring algorithms, researchers report.
- Artificial Intelligence
AI-designed proteins test biosecurity safeguards
AI edits to the blueprints for known toxins can evade detection. Researchers are improving filters to catch these rare biosecurity threats.
- Genetics
AI generated its first working genome: a tiny bacteria killer
Bacteriophages designed with AI kill E. coli faster than a well-studied strain, but the tech needs regulation before moving beyond lab dishes.
- Health & Medicine
Can AI spot harmful health side effects on social media?
A new AI tool discovers harmful side effects of cannabis products from Reddit posts. Public health workers could use this info to help keep people safe.
- Tech
This experimental computer chip reuses energy
A first-of-its-kind test shows that reusing energy within a computer chip can work, thanks to two techy tricks.
