11/05/2024 | News release | Distributed by Public on 11/06/2024 06:44
One of the standout themes at BioTechX 2024 was the real progress in AI, especially Generative AI. Last year, we were all about the potential of large language models and generative AI. This year, it was all about real-world applications and measurable impacts. For example, a keynote presentation showcased a generative AI platform that's been scaled up, boosting efficiency by 10-20% for its 15,000 users. This shift from theoretical potential to practical implementation was a recurring theme throughout the event.
As organizations continue to explore AI applications, the emphasis on trustworthy data has never been greater. Several sessions highlighted that the success of AI models in life sciences heavily depends on reliable data. While defining and obtaining trustworthy data is challenging, there were successful examples where robust data foundations led to significant advancements. The consensus was clear: without trustworthy data, AI models can't reach their full potential.
Another hot topic was multi-agent systems. It's clear that the future of AI in life sciences involves multiple specialized AI models working together. This approach, known as agentic AI, uses a suite of models, each with its own specialization, to tackle complex scientific workflows. This method not only enhances AI efficiency but also addresses the unique challenges of life sciences research.
Discussions also touched on the need for pragmatic approaches to FAIR (Findable, Accessible, Interoperable, and Reusable) data. While everyone agrees on the importance of FAIR data, achieving perfect FAIRness isn't always practical. The key takeaway was that data should be FAIR enough to be functional and useful, without overcomplicating the process.
Stay tuned for our next blog, where we'll dive deeper into our panel discussion on FAIR data.
Quantum computing was another exciting topic at the event. While it is still seen as a technology with a 5 to 10-year horizon, its potential to revolutionize drug discovery was widely discussed. Quantum computing is expected to solve problems that current technologies cannot, offering new ways to understand and manipulate biological systems. However, it was also acknowledged that quantum will be more complementing rather than replacing existing technologies. One takeaway is that we have to think about how to combine classical technologies and high-performance computing (HPC) cleverly with quantum technologies and think out of the box to achieve revolutionary advancements.
In addition to its potential, the event highlighted various use cases and benchmarks of existing quantum algorithms and hardware. These discussions underscored the significant hardware development needed to address interesting use cases in the pharmaceutical industry. Nevertheless, it is crucial to start now. Beyond hardware advancements, there is a substantial amount of work required on the implementation side, such as developing effective error correction strategies. On the software side, optimizing quantum algorithms is essential to achieve these goals as soon as possible.
No it's not new, but a recurring topic at BioTechX 2024 was the ongoing productivity crisis in R&D. Despite increased spending, the number of new drug approvals has not kept pace (Research and development costs, currently exceeding $3.5 billion per novel drug, reflect a five-decade decline in pharmaceutical R&D efficiency.). The event highlighted the need for innovative solutions to boost productivity while reducing costs. AI and other advanced technologies were seen as crucial to achieving these goals, with a focus on practical applications that can deliver measurable results.
In conclusion, BioTechX 2024 was a pivotal event that highlighted the shift towards realism and practical applications in the biotech and life sciences industries. The discussions underscored the importance of trustworthy data, the potential of multi-agent systems, and the future role of quantum computing. As the industry continues to evolve, these insights will be crucial in guiding future advancements and addressing the challenges ahead.
We are sorry, the form submission failed. Please try again.