Can AI “co-scientists” help answer cancer’s biggest questions?
Three teams of cancer researchers are testing whether AI co-scientists can help turn clues buried in scientific papers and research data into new ideas for experiments. The post Can AI “co-scientists” help answer cancer’s biggest questions? appeared first on Cancer Research UK - Cancer News .
Three research teams are testing whether artificial intelligence (AI) can assist in cancer research by serving as virtual lab assistants. The £8m AURORA project, led by Cancer Grand Challenges, aims to tackle the biggest and most important questions in science using AI. A specialized AI platform called AURORA's "co-scientists" will be evaluated by the teams as they investigate various aspects of cancer.
These AI co-scientists cannot make scientific decisions independently but can analyze vast amounts of data, generate new ideas based on hidden patterns, and prioritize promising experiments. They are designed to stick to real scientific evidence and recognize when information is uncertain. Unlike AI chatbots, which often "hallucinate" fake answers, AURORA's co-scientists draw from properly reviewed scientific resources and research data. When data is unavailable, they will acknowledge the limitation.
The technology, called "agentic" AI, allows the co-scientists to guide themselves through tasks step by step with autonomous reasoning powers. Researchers can still have control and oversight, but they won't have to spend all day at the keyboard, as the AI can comb through vast libraries and databases without constant supervision. The ultimate goal is to give researchers more time to focus on aspects requiring their expertise.
The three research teams will use AURORA to test AI's potential in answering questions related to cancer. The NexTGen team will explore how immune cells can be modified to better target solid tumours in children and young people. The PROMINENT team will investigate why some cells with cancer-linked genetic changes remain healthy while others turn cancerous, considering factors like age, sex, and exposure to cancer-causing substances.
The MATCHMAKERS team will use AI to explore how T cells recognize cancer, seeking the most promising markers on cancer cells and immune cells to attack those cells, potentially leading to more personalised treatments.
In conclusion, AURORA's co-scientists could potentially bring significant benefits to cancer research by speeding up the discovery process and helping uncover new insights. However, the final experiments and validation of AI's suggestions remain crucial, underscoring the importance of human expertise in the scientific process.
Written by urgent.news from Cancer Research UK's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
