Add BrightSurf on Google Email

Binghamton University


Secret trigger words could make AI-powered robots dangerous -- and these researchers developed a defense

Researchers at Binghamton University have demonstrated that hidden trigger words inserted into LLMs can cause robots to go off-script and perform dangerous actions. To combat this, they developed a defense mechanism called PROTEA, which evaluates task plans for signs of an attack and prevents harmful actions from taking place.

SourceBinghamton University·TypeExperimental study·DateOct 20, 2026

When AI chatbots give bad advice, no one can see the damage

A new paper explores how chatbot health guidance leaves no trace for clinicians, regulators, and users, highlighting the need for transparency safeguards. The authors propose methods to increase transparency and responsibility for AI companies, including access to user conversations, disclosure systems, and third-party monitoring.

SourceBinghamton University·JournalNature Health·TypeCommentary/editorial·DateOct 9, 2026

Binghamton University’s Innovative Simulation and Practice Center named a SUNY Nursing Simulation Center of Excellence

Binghamton University's Innovative Simulation and Practice Center has been named a SUNY Nursing Simulation Center of Excellence, receiving a $10 million investment to expand simulation space and experiences. The center will focus on faculty training and professional development in simulation-based education, addressing gaps in clinical...

Study: Sibling interactions are an untapped resource for early engineering education

Researchers at Binghamton University identified clear roles among siblings on STEM tasks, including colleague, lead engineer, and provocateur roles. Sibling relationships offer a safe space for children to develop engineering skills, and programs should tap into this relationship for effective education.

SourceBinghamton University·JournalJournal of Engineering Education·TypeObservational study·DateSep 15, 2026

S-M-A-R-T! These researchers used math to crack Wordle

A team of researchers at Binghamton University developed a method to solve Wordle with a 99% success rate by applying information theory. By prioritizing guesses that provide the most information, they reduced the pool of possible words and solved puzzles in fewer guesses.

SourceBinghamton University·JournalNortheast Journal of Complex Systems·TypeComputational simulation/modeling·DateJun 17, 2026

AI without hallucinations: Binghamton University researchers develop new way to reduce troublesome fake info

Binghamton University researchers have developed a new way to reduce troublesome fake information in AI chatbots, with high accuracy in identifying disease terms and drug names. The protocol harnesses multiple large language models to verify answers through 'voting', increasing confidence in the results.

SourceBinghamton University·JournalSTAR Protocols·TypeComputational simulation/modeling·DateJun 2, 2026