Flint Global

It was a great pleasure to be invited by GSMA APAC to dinner with industry and government figures and to then moderate a panel on scams at GSMA's Digital Nation Summit Manila. I was struck by the dedication I found towards using digital technologies to advance inclusion and raise living standards in the Philippines. The threat of scams is one of the biggest obstacles to this ambition because of the damage it does to trust in digital services. On the panel, Justice Reyes Jr. of the Cybercrime Investigation and Coordinating Center (CICC), John Gonzalez of PLDT Enterprise, and Lito Villanueva of RCBC discussed how government and different sectors are collaborating to address the threat. This includes through both developing tools such as the use of device location to flag suspicious transactions and through implementing educational initiatives to inform both users and businesses on the threat.

💥 NEW PAPER: A Comprehensive Analysis of AI Regulation in India💥 "What's happening on AI regulation in India? What should the government do?" If these questions pique your interest. I'd like to share a new paper I've published with Carnegie India. It captures the general sentiments on AI regulation and suggests a policy roadmap for India based on extensive consultations and research. Some of our findings were surprising — turns out government, industry and civil society agree on many aspects of AI regulation. That makes me hopeful that we can convert consensus into action. But there are some roadblocks — though we have some intuitions about the novel risks of AI, we don't yet have empirical evidence of harm in the Indian context to frame new regulations. There's also the complex task of figuring out where the gaps lie in existing laws, and how to plug them in efficient ways. Having worked in technology law for over a decade, I can tell you this is no mean feat (see Part 2). Another concern is the lack of state capacity. Expertise and enforcement is a challenge. We need to think up agile, flexible and innovative frameworks to supplement current capacity (an AI Safety Institute for India perhaps?). Overcoming these obstacles will require a multi-fold strategy. For a country like India, we recommend a targeted and iterative approach that blends capacity building, risk assessments and self regulation with some government oversight (see Part 4 of the paper). We also identify key agencies that should be involved in this process. This is the first comprehensive analysis of AI regulation in India, so we welcome your feedback and hope you find it informative, accessible and useful. Finally, I have two requests: 1. Please share this paper with anyone interested in AI policy - if nothing else, the 100+ footnotes are a treasure trove of resources.  2. If you're looking for someone from India to speak to a global audience about AI policy, please reach out. We want to ensure that AI regulations are interoperable and that we are engaged with our counterparts around the world. As always, many thanks to my co-author Shatakratu, the anonymous interviewees and reviewers, and to the entire Carnegie India team that helped put this paper together!

It was a very upbeat vibe at the British Screen Forum 2024 Conference this week, “where the best informed and most influential people in the UK screen sectors convene to interrogate issues of importance and influence policy and the thinking around policy". In one session Dame Melanie Dawes, Chief Executive of Ofcom, was interviewed by Ed Richards, formerly in that role and now Managing Partner of Flint Global, about ‘Ofcom at 21’. Some key takeaways for me were: 👉Ofcom’s overriding objective has always been and remains ‘Making communications work for everyone’ 👉New areas of responsibility for Ofcom (in particular online safety) are getting extra funding so will not detract from existing areas such as broadcasting (funding is ringfenced by area) 👉Although there have been big changes in people's viewing habits (such as a major switch from TV to social media as a source of news and entertainment particularly in the under 30s), they are still consuming a large amount of amazing British content 👉The Media Act updates the regulatory framework in relation to PSBs, and there is now a follow up review as to what more needs to be done given the seismic changes in this area, looking at the sustainability of PSB funding models whilst preserving the unique and important role played by PSBs in the UK's media ecosystem. 👉The OSA is going to impact on many businesses, big and small, including in previously less regulated areas like gaming. The next big batch of draft Codes of Practice will be published in mid December and finalised by next Spring so that everyone knows what they have to do, and Ofcom will expect to see action in achieving compliance 👉Safety is a key issue under the OSA, in particular risk assessments and age verification, with a focus on anti-grooming measures and reduction of harmful content such as violence, suicide/self harm and pornography for under 18s 👉Dame Dawes didn’t “read the situation” that the plan was for Ofcom to “regulate AI”, and she didn’t know what that really meant. Before regulating anything you needed to know what harm you were trying to prevent and what services would be in scope. At present, where activities are within existing regulatory frameworks they would be regulated irrespective of the technology 👉Ofcom’s current work on AI was on the basis of existing areas of responsibility, e.g. use of Gen AI to better identify telecom scams and in broadcast to address risks of misinformation