Dylan Pahina’s Post

https://lnkd.in/diHRA6pW Fostering AI Innovation and Competition: The Way Ahead Authors: Ettore Greco 03/10/2024 The intensifying dialogue and cooperation between competition authorities across the Atlantic and in the G7 framework to deal with risks stemming from the rise of generative AI is a promising development. A convergence of efforts and policies to safeguard market competition and the interests of consumers can greatly contribute to countering the disruptive effects of rapid technological change and building transparency and trust. The AI markets are however at an early stage of their development and the future trajectory of AI technologies and their multi-faceted applications remains highly uncertain. This requires a thorough and continuously updated study of market dynamics. While there is a rising feeling that early action by governments and regulators is needed to counter anticompetitive drives, premature interventions run the risk of stifling competition. Dialogue with market players and stakeholders is of paramount importance to identify the emerging competition risks and evaluate the impact of the existing pro-competition tools or new regulatory measures on the AI realm. The outreach activities undertaken by governments and competition authorities provide a solid basis for further advancements in the dialogue with stakeholders. Voluntary codes of conduct such as the one launched by the Biden administration and the G7 may play a role in promoting responsible behaviour by industrial actors. In this context, the G7 can provide an essential platform for advancing discussions on the role of competition policy in AI markets to ensure adequate actions to address anticompetitive issues. The G7 has also proved to be a valuable instrument to foster policy convergence among its members on competition rules and more broadly governance models for generative AI.

𝗔 𝗖𝗘𝗢’𝘀 𝗚𝘂𝗶𝗱𝗲 𝘁𝗼 𝗟𝗲𝗮𝗱𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜 Generative AI is transforming the business landscape. Our recent survey with Fortune/Deloitte reveals that 79% of CEOs see accelerating innovation as a key benefit of adopting Generative AI. As a CEO, your direct involvement in Generative AI initiatives is crucial. It’s important for you, as a leader, to understand the opportunities and challenges of Generative AI. Here’s how you can lead effectively: ✔ 𝐌𝐚𝐤𝐞 𝐒𝐭𝐫𝐚𝐭𝐞𝐠𝐢𝐜 𝐃𝐞𝐜𝐢𝐬𝐢𝐨𝐧𝐬: Decide if your company will be an early adopter (a first mover) or if it will quickly follow new trends (a fast follower). Consider whether Generative AI will be used for innovating new products or for improving productivity. ✔ 𝐔𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝 𝐭𝐡𝐞 𝐓𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐲: Even if you are not a tech expert, having a basic understanding of how Generative AI works will help you make informed decisions. Engage with your CIO and CTO to align technology choices with your business goals. To leverage Generative AI, having the right 𝐭𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐢𝐜𝐚𝐥 𝐢𝐧𝐟𝐫𝐚𝐬𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞 is essential. ✔ 𝐂𝐡𝐨𝐨𝐬𝐞 𝐒𝐦𝐚𝐫𝐭 𝐈𝐧𝐯𝐞𝐬𝐭𝐦𝐞𝐧𝐭𝐬: The demand for AI chips is growing, and it’s important to secure the necessary computing power. Discuss with your tech leaders about investing in specialized AI chips that can handle large amounts of data efficiently. ✔ 𝐄𝐯𝐚𝐥𝐮𝐚𝐭𝐞 𝐀𝐈 𝐄𝐜𝐨𝐬𝐲𝐬𝐭𝐞𝐦𝐬: Select an AI ecosystem that fits your company’s needs. This might mean choosing to build a private LLM to keep control over your data and processes, or partnering with established AI providers for quicker implementation. Creating 𝐀𝐈 𝐜𝐞𝐧𝐭𝐞𝐫𝐬 𝐨𝐟 𝐞𝐱𝐜𝐞𝐥𝐥𝐞𝐧𝐜𝐞 within your company can accelerate the effective use of Generative AI. ✔ 𝐅𝐨𝐬𝐭𝐞𝐫 𝐈𝐧𝐧𝐨𝐯𝐚𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐆𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞: These centers should bring together experts from various fields to focus on developing AI tools and ensuring they are used responsibly and ethically. Make sure these centers have the resources they need to succeed. ✔ 𝐏𝐫𝐨𝐦𝐨𝐭𝐞 𝐀𝐈 𝐔𝐧𝐝𝐞𝐫𝐬𝐭𝐚𝐧𝐝𝐢𝐧𝐠 𝐚𝐧𝐝 𝐀𝐝𝐨𝐩𝐭𝐢𝐨𝐧: Encourage a culture where employees are informed about AI and its benefits. This helps in managing changes and ensures that AI tools enhance employee capabilities rather than replace them. Our Thought Leadership Series is designed to help you navigate the complexities of Generative AI: https://lnkd.in/dVAYRtux 𝐋𝐞𝐚𝐝𝐢𝐧𝐠 𝐰𝐢𝐭𝐡 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐢𝐬 𝐧𝐨𝐭 𝐣𝐮𝐬𝐭 𝐚𝐛𝐨𝐮𝐭 𝐚𝐝𝐨𝐩𝐭𝐢𝐧𝐠 𝐧𝐞𝐰 𝐭𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐲 - 𝐢𝐭’𝐬 𝐚𝐛𝐨𝐮𝐭 𝐬𝐡𝐚𝐩𝐢𝐧𝐠 𝐭𝐡𝐞 𝐟𝐮𝐭𝐮𝐫𝐞 𝐨𝐟 𝐲𝐨𝐮𝐫 𝐜𝐨𝐦𝐩𝐚𝐧𝐲 𝐰𝐢𝐭𝐡 𝐢𝐧𝐟𝐨𝐫𝐦𝐞𝐝, 𝐬𝐭𝐫𝐚𝐭𝐞𝐠𝐢𝐜 𝐝𝐞𝐜𝐢𝐬𝐢𝐨𝐧𝐬. Are there any crucial aspects of Generative AI that you think we should discuss more? Share your thoughts in the comments below! 𝐶𝑟𝑒𝑑𝑖𝑡𝑠 𝑡𝑜 𝑃𝑢𝑓𝑓 𝑌𝑎𝑐ℎ𝑡𝑦

Takeaway 1: Subjectivity towards AI can be challenging for global innovation Last week I attended the "Responsible AI in Global Business" conference at the Yale School of Management and had the pleasure of meeting Jon Iwata who hosted the event. In every session, there was a mixture of information, insight and opinion. What was most evident to me was the imbalance of knowledge and varying priorities everyone had. This is to be expected, the panelists and speakers all have exceptionally different backgrounds. There was unilateral agreement that #AI could be a very good thing. Many of the panelists in different sessions did end up in various states of disagreement about governance, use cases, model types, and AI impacts. This was the most eye opening learning from the event. If one leader believes in relentless #innovation and another focused on tightly constrained use cases driving new model development they will have different points of view on the use of AI, and how to govern it. If there is an advocacy for broad models designed for many tasks and a competing methodology for many small scale models designed for individual tasks, their strategies to infuse AI into their businesses will be in opposition. If half of the global community believes in #OpenSourceAI and the other half is developing commercialized offerings, can there be normalized research and training? The split even extended to governance of AI and the data marketplaces to train and store outputs from models. Mainstream AI rollout is still very much in its infancy and because there are so many varying and at times opposing points of view, duplicative efforts are inevitable. After a spirited discussion with a colleague following the event I realized the observation around subjectivity had to be the first key takeaway. Objectively, everyone acknowledges that generative AI can pose a risk to humans because in the wrong hands it can create damaging disinformation or produce outcomes which can be harmful. Subjectively, some people are so keen on seeing just how far it can go that models are built for the sake of building them with little or no planning for guardrails. Objectively, we all agree that laws and rules are in place for a reason and should be obeyed. Subjectively, groups are in favor of decentralized AI regulations and limited restrictions so as not to hamper rapid new development. Objectively, it makes sense to focus effort, time and money on specific business needs or goals. Subjectively, AI research teams are placing attention on being first to innovate or loosely seeing how far a project can be taken before hitting a barrier. Businesses, researchers and academics should work to reach agreement on the topics of innovation, applicability and #governance on a global scale. I believe that's the best way to promote innovation without complicating adoption for business. If and when that will happen is the unanswered question.

There are many articles and opinions on how the adoption of AI is moving along. Thank you to Oliver King-Smith of smartR AI for sharing his insight on this subject in my knowledge base. _______________________ "While embracing AI is crucial for staying competitive to keep up or leapfrog its competitors, overzealous adoption without proper experience can lead to detrimental outcomes. These companies must prioritize gaining experience and proficiency in implementing and managing AI solutions effectively before attempting widespread integration." ________________________ For other articles and opinions, visit our collection - AI Adoption and Usage: https://lnkd.in/gHXa3sz3 ______________________ #ai #aiethics #chatgpt #LLM #automation #aiadoption Erica W Andersen

This October, we strutted our stuff at the World AI Summit 2024 in Amsterdam, rubbing shoulders with the who's-who of artificial intelligence. The goal? To show the world (and a lot of curious business leaders) that Kosmoy is the GenAI powerhouse that enterprises have been waiting for. What Did We Bring to the Party? 1. AI Governance, But Make It Smart AI adoption in companies can get messy fast—think Wild West vibes. We've got this covered with our centralized AI governance, keeping things in check across all departments. It’s like a bouncer for your AI apps, making sure only the good ones get in. Plus, we’ve got role-based access management (RBAC) so everyone knows their lane, and we even help you comply with the EU AI Act (no, it’s not optional, and yes, we make it easy). 2. Keeping an Eye on AI You can’t just let your AI go rogue, can you? Our observability tools keep tabs on performance, flag anything toxic (no bad vibes here), and track how much these models are costing you. Think of it as the ultimate AI babysitter that never takes its eyes off the prize. 3. Building AI? We’ve Got Your Back Need to build a chatbot that actually understands your corporate lingo? Our “RAG-in-a-box” (Retrieval Augmented Generation, but let’s stick with RAG) is like a starter kit for smart chatbots. Plus, we’ve got some slick tools to handle unstructured data (graphs, tables, you name it), track changes to prompts, and integrate effortlessly with all the big LLMs like OpenAI and LLaMA. The Takeaway? At the World AI Summit, we didn’t just sit at the table—we showed up with solutions. Our pitch? Fast, scalable, and totally manageable AI for enterprises. If your company is stuck in AI chaos, we’re like the tech superhero that swoops in, brings order, and makes sure your AI systems work for you, not against you.

A simple summary of what's needed.

The #AI Conversation: Why We're Not Going All-In (Just Yet) 🤔 The buzz around AI is louder than ever—and with good reason. This powerful technology has the potential to transform industries and redefine what's possible. Big tech is deeply invested in AI research and development. They're integrating AI solutions across their operations and seeing impressive results. However, we're not going "all-in" on AI just yet. Here's why. ✅ 1. Building Trust Takes Time AI is still an emerging field, and establishing trust is essential. We've all witnessed instances of AI bias, lack of explainability, and occasionally unsettling outputs. Simply applying AI to every problem isn't the answer. We need strong ethical frameworks to ensure our AI is fair, unbiased, and used responsibly. This involves collaboration with ethicists, policymakers, and the public—a process that requires time and dedication. ✅ 2. Humans in the Loop Let's not overlook the human element. AI is a remarkable tool, but it doesn't replace human ingenuity, creativity, and empathy. We believe in the synergy of human-AI collaboration. AI can automate tasks, analyze vast datasets, and uncover patterns that humans might miss. But it's humans who interpret these insights, make final decisions, and ensure the technology serves a beneficial purpose. ✅ 3. The Need for Explainability "Black box" AI—where data goes in and an answer comes out without understanding why—isn't sufficient. We need explainable AI that reveals its decision-making process. This transparency is crucial for building trust, identifying and mitigating bias, and ensuring regulatory compliance. As Elon Musk recently emphasized, "It's very important that AI be trained to be truthful and not trained to be politically correct." ✅ 4. Prioritizing Human-Centered Design AI shouldn't be an afterthought bolted onto existing systems. The industry should prioritize human-centered design and involve people throughout the AI development process—from ideation to deployment—to ensure the technology serves and empowers users. ✅ 5. The Long Game While some companies chase quick headlines, the best strategy is to focus on the long game: building a robust AI infrastructure, fostering a culture of responsible innovation, and attracting top AI talent. This approach takes time and strategic investment. It's better to build it right than to rush a half-baked solution to market. This doesn't mean to stand still. Big tech is actively developing exciting AI projects that are making a positive impact. Our challenge is doing so responsibly, with a clear understanding of the challenges and limitations. The future is undoubtedly AI-powered, but it’s a future that must be built with careful consideration, human oversight, and a commitment to ethical development. Let's continue this important conversation. What are your thoughts on AI's role in the future? How can we ensure it's used for the greater good❔

The Year 2022 was a watershed year for artificial intelligence, with the release of several consuming-facing applications like ChatGPT. Dall.E. and Lensa. [1] The common theme is the use of Generative AI-a paradigm shift in the world of AI. While current generations of AI use pattern detection or rule following to help analyse data and make predictions, the advent of transformer architectures[2] has unlocked a new field. Generative artificial intelligence.[3] Generative AI can mimic the human creative process by creating novel data similar to the kind that it was trained on, elevating AI from enabled to potentially co-passenger.[4] Although early traction has been through consumer release, which could be era defining, Generative AI has the potential to add contextual awareness and human like decision making to enterprise workflow, and could easily change the way we do business. We may only just be beginning to see the impact of solutions like Google’s contact centre which is designed to enable Natural language customer service interactions, and industry specific solutions. In 2022 venture capital forms have invested $2Billion and technology leaders made significant investment, such as Microsoft’s $10Billion stake in Open AI and Google’s $300 million stake in Anthropic.[5] Still questions remain about how engineers and engineering enterprises could use generative AI to deliver design and efficiency gains, product improvements or operational change. Similarly we are only beginning to see how Generative AI could be commercialised and how to build a sustainable business models. Even so, Generative AI is in its infancy and not without risk. Ultimately Generative AI could create a more profound relationship between humans and technology, even more than the cloud, the smartphone, and the internet did before.[6] The AGI “market” will likely double every two years for the next decade.[7] Numbers aside we believe the economic impact could be far greater. [1] https://lnkd.in/gBb7gkXT [2] https://lnkd.in/gDcek5jU) [3] https://lnkd.in/g3Vyh8EQ [4] https://lnkd.in/gK3T5sqj [5] https://lnkd.in/dr4pm8Uz [6] https://lnkd.in/gc8zVejA [7] https://lnkd.in/ejyQ2aUb

When it comes to investing in #AI revolution, distinguishing true winners from the hype can be challenging. I use following criteria to find sustainable AI 🏆 1. Corned AI resource: Borrowing a concept from Hamilton Helmer's 7 Powers, we arrive at 1st AI-era success factor: cornered resource. Helmer defines as preferential access, under attractive terms, to coveted asset that can independently enhance value. CO’s with preferential access to coveted AI assets, such as proprietary data, specialized AI h/ware, or a 🧠trust have a powerful competitive edge when kept under exclusive control. 2. AI Driven Differentiation: To find investable AI CO’s capable of delivering premium revenue growth expected from their substantial CAPEX, focus on those applying AI in truly differentiated ways AI driven differentiation about leveraging AI to set apart from the comp. by delivering unique, high-impact value. AI-driven differentiation happens when AI is applied to solve complex problems, create 🆕 experiences, or innovate that’s hard for others to match 3. Adaptive AI Excellence: History shows early adaptation during major platform shifts is crucial, as initial stages are filled w/ uncertainty. In AI, unknowns persist: Will LLM advancements hit limit? Can synthetic data improve models as real 🌎 data diminishes? CO’s must navigate evolving reg, privacy, & security risks. Long-term success may depend on unforeseen breakthrough, or on overcoming fundamental limitations. To thrive, companies need to continuously reassess their strategies, pivot, & adapt to ever-changing AI landscape. Adaptive AI Excellence is all about a company's ability to not only build cutting-edge AI systems but also remain flexible and responsive as AI ecosystem shifts. Balance between pushing frontiers of building AI tech & staying agile to adapt quickly to new players; limitations in AI models, UI & h/ware; 🆕 biz models & innovations like OpenAI o1. 4. 💪 leadership: visionary leaders can make the difference between a CO that promises AI transformation & one that advances AI and wields its power to create long-term value. Leadership in AI is not just about technical expertise; it’s about having the foresight to navigate a fast-moving landscape, the courage to make bold decisions, and the ability to inspire and align teams around a clear vision for AI’s role in the CO’s future. 5. Parallel AI Bets: To be used when evaluating megacap’s as potential future AI winners, focus on co’s making parallel AI bets. Large co’s investing in multiple AI projects/tech simultaneously helps them hedge against uncertainty of fast-moving AI ecosystem by spreading $$ across range of innovations. way for co’s that can afford it to increase their prob. of success. Co’s can place multiple bets on different AI tech, develop proprietary AI models while partnering externally, or target both comm. & consumer markets https://lnkd.in/gQeWBHJG

🤖 “By now, it’s evident that #AI will change everything. But, as with previously transformative #technologies, it will be access to this #technology that defines its #impact on the world. Henry Huang, CTO of NetMind.AI, explores the importance of democratising AI. https://lnkd.in/gE9h5uyr