Large Reasoning Models like DeepSeek-R1 mark a fundamental shift in how LLMs approach complex problems. Instead of directly producing an answer for a given input, DeepSeek-R1 creates detailed multi-step reasoning chains, seemingly ‘thinking’ about a problem before providing an answer. This reasoning process is publicly available to the user, creating endless opportunities for studying the reasoning behaviour of the model and opening up the field of Starting from a taxonomy of DeepSeek-R1’s basic building blocks of reasoning, our analyses on DeepSeek-R1 investigate the impact and controllability of thought length, management of long or confusing contexts, cultural and safety concerns, and the status of DeepSeek-R1 vis-à-vis cognitive phenomena, such as human-like language processing and world modelling. Our findings paint a nuanced picture. Notably, we show DeepSeek-R1 has a ‘sweet spot’ of reasoning, where extra inference time can impair model performance. Furthermore, we find a tendency for DeepSeek-R1 to persistently ruminate on previously explored problem formulations, obstructing further exploration. We also note strong safety vulnerabilities of DeepSeek-R1 compared to its non-reasoning counterpart, which can also compromise safety-aligned LLMs.
Artificial Intelligence, particularly Generative AI, is making a significant impact on India’s insurance industry in 2025, delivering productivity improvements of over 30 per cent, according to a new report by Boston Consulting Group (BCG) released on Friday.
The report highlights that insurers leveraging AI in underwriting are witnessing efficiency gains of up to 36 per cent, driven by improved use of both structured and unstructured data. In customer service, AI-enabled knowledge assistants have enhanced productivity by more than 30 per cent, while simultaneously raising service quality standards.
AI is also strengthening claims processing. The report notes that up to 70 per cent of straightforward claims are now being resolved in real-time using AI, reducing processing costs by 30–50 per cent and significantly improving the customer experience.
Ora Biomedical, in partnership with Rapamycin Longevity Lab, announces the successful funding of the first subproject under its ambitious initiative to conduct a rapid lifespan analysis of 601 mTOR inhibitors in roundworms.
With $50,000 secured, Ora Biomedical will now commence the next phase of the first subproject. This will be a high-throughput screening of 301 mTOR inhibitors using its cutting-edge WormBot-AI technology. This milestone marks an important step toward identifying next-generation compounds that could be more effective than rapamycin, which is currently seen as the golden standard because of its good longevity effects in multiple species.
Mitchell Lee, CEO of Ora Biomedical, emphasized the importance of this research by stating: “The potential of targeting aging to broadly improve healthy lifespan is clear from decades of studies with compounds like rapamycin. However, even for well-validated molecular targets like mTOR, we still don’t know the best interventions. We at Ora Biomedical are proud to partner with Rapamycin Longevity Lab to advance our understanding around targeting mTOR and related kinases for maximizing healthy lifespan. None of this work is possible without support from visionary donors and organizations like the Lifespan Research Institute, the nonprofit behind Lifespan.io, with whom we have partnered to create pathways for donations to advance longevity science. To all those involved, thank you again, and we are excited to get to work!”
Silicon Valley startup Lightmatter has developed a novel computer chip that can speed up artificial intelligence processes and save electricity in the process. The company focuses on using beams of light to move data between computers rather than using electric signals.
Connection speeds are a great matter of concern when it comes to artificial intelligence due to its complex software. This complexity requires the software to be spread over many computers.
The UK is developing a controversial AI tool to predict potential murderers using personal data from convicts, victims, and witnesses. Aimed at preventing crime, the initiative mirrors global trends in AI policing. However, critics warn of bias, privacy issues, and call the project “chilling and dystopian,” raising ethical concerns about pre-crime technology.
UK | artificial intelligence | AI | murder | firstpost | world news | news live | vantage | palki sharma | news.
#uk #ai #murder #firstpost #vantageonfirstpost #palkisharma #worldnews.
Vantage is a ground-breaking news, opinions, and current affairs show from Firstpost. Catering to a global audience, Vantage covers the biggest news stories from a 360-degree perspective, giving viewers a chance to assess the impact of world events through a uniquely Indian lens.
The show is anchored by Palki Sharma, Managing Editor, Firstpost.
