Luminos.AI

?? Generative AI and Copyright: Why Current Legal Frameworks Fall Short — and What Companies Should Do ?? Existing copyright law in the U.S. and text-and-data mining (TDM) exceptions in the EU don’t fully address the use of massive datasets to train GenAI models - meaning that many companies can feel like they are flying blind when it comes to clear legal rules for their GenAI training data. Neither U.S. fair use doctrine nor EU TDM exceptions were designed with foundation models in mind — leaving companies exposed to uncertain, untested legal risks. Professors Tim W. Dornis and Sebastian Stober provide a great overview of these issues in a new paper, Generative AI Training and Copyright Law (link below). ?? Why this matters for companies: Legal, privacy, and AI governance teams are already overwhelmed with GenAI approvals. Without knowing what data a model was trained on — or how that data interacts with evolving copyright and privacy law — approving AI tools is a minefield. Manual review is impossible at scale. ?? The way forward? ? Automated training data governance solutions to evaluate datasets before models are deployed. ? Proactive AI governance processes that align with fast-evolving global regulations. And regulation is coming fast: ?? The EU AI Act mandates training data transparency for high-risk models. ?? The FTC has warned AI trained on unlawful data may violate consumer protection laws. ?? The UK has called for clear data accountability in AI development. Bottom line: If you’re building or buying GenAI tools, understanding and managing training data risks is no longer optional. Want to learn more about automated approaches to approving GenAI systems for risks? Reach out to [email protected]! #GenerativeAI #AIgovernance #Copyright #TDM #AIregulation #InHouseCounsel #AIethics Read the paper here: https://lnkd.in/eESuEy4G

Enterprises are betting big on AI—but getting models approved for legal risk has become their biggest challenge. And this will only get worse with laws like the EU AI Act, the Colorado and Utah AI Acts, and more. ? The AI approval process is slow, complex, and often ad hoc, delaying innovation. ??? Our CEO Andrew Burt sat down with Ben Lorica 罗瑞卡 on The Data Exchange Podcast to unpack why legal hurdles are holding AI back—and how to fix it. ? At Luminos, we’re streamlining AI adoption by automating AI approvals—so your models move from idea to production, faster without getting stuck in an approval process bottleneck. ?? Listen here: https://lnkd.in/gYr7A2K8 ?? Want to see how Luminos can help? Email us at [email protected].

What’s the biggest challenge posed by artificial intelligence? It’s not social media, the spread of misinformation, fears of Skynet, or all the other harms that dominate the headlines. Instead, it’s much simpler. The problem is that organizations adopting AI do not have the right *tools* to manage its risks. Data scientists and legal teams live in different worlds, use different tools, and have an extremely difficult time communicating. And this means that companies have a choice: adopt AI and overlook compliance, or emphasize compliance and jeopardize the speed at which they adopt AI. The way to solve for this problem is to give these teams the tools they need to communicate, test and manage AI risks efficiently and at scale. In short, to automate the way that AI is managed for risks. And this is exactly what we’re focused on at Luminos. Read our latest on the disconnected between lawyers and data scientists below, and please reach out to us at [email protected] if you're interested in learning more! https://lnkd.in/eEUsAP8K

What do you need to know about AI in 2025? Ben Lorica 罗瑞卡 provides the key highlights over at Gradient Flow - describing the need for platforms that *align* testing and tools across AI risks. That's because in practice, legal and data science teams have too many different tools for workflows, risk triage, testing, documentation and more - which is why it can take months to oversee and approve AI systems for risks. But all of that changes with AI alignment platforms like Luminos.AI. Read more about AI alignment platforms below - and reach out to [email protected] to learn more! https://lnkd.in/eRwb8CAZ

Congrats to the entire Luminos.Law and ZwillGen team! Compliance has become *the* biggest barrier to AI adoption for every company that is serious about AI. With our roots in the legal and privacy community - and Luminos.Law in particular! - we know that it takes a creative, interdisciplinary team to allow enterprises to move fast and ensure AI compliance. The new AI Division at ZwillGen will be a central part of that approach! To learn more about our tools for automated AI oversight, reach out to us at [email protected]. We'd love to talk to you! https://lnkd.in/epvG7hFw

Is this your AI governance solution: some combination of documents, spreadsheets, email, and ad-hoc meetings? The state of the art in GenAI is evolving rapidly--and so is the legal landscape you'll need to comply with. Your legal, technical, governance, and business teams need to collaborate closely to understand the risks, complete any needed testing and documentation, and get the necessary sign-offs. But if you're like most organizations we talk to, this AI governance process is a hodge-podge of disconnected tools, delayed responses, confusion and furstration. The result? It often takes 3-6 months or more just to get deployment approval for a new AI tool, even after development is done! It doesn't have to be that way.?Luminos.ai's Workflow Manager, part of our AI Alignment Platform, is designed specifically to bring these teams together in one collaborative environment. One where you can build on industry best practices, easily construct your own workflows and governance documents, and get input from all the internal & external stakeholders--rapidly and with full transparency. If you're interested to take your AI governance approval process from 3-6 months down to 3-6 days, reach out to?Luminos.ai?at [email protected] or visit our website??? https://www.luminos.ai/

Misuse and harms from Generative AI are no longer theoretical problems: they are happening today. The article below explores recent research from DeepMind on the topic -- recommended reading for anyone interested in the space. If you are looking to strengthen your own Responsible AI initiative and mitigate these risks, our alignment platform can help!?How??By combining easy collaboration between legal, data science, and business teams with rigorous testing and clear documentation. Interested in learning more? Reach out at [email protected] or visit our website??? https://www.luminos.ai/ #GenerativeAI #legal #datascience

Misuse and harms from Generative AI are no longer theoretical problems: they are happening today. The article below explores recent research from DeepMind on the topic -- recommended reading for anyone interested in the space. If you are looking to strengthen your own Responsible AI initiative and mitigate these risks, our alignment platform can help!?How??By combining easy collaboration between legal, data science, and business teams with rigorous testing and clear documentation. Interested in learning more? Reach out at [email protected] or visit our website??? https://www.luminos.ai/ #GenerativeAI #legal #datascience

The ?? AI Risk Repository consolidates insights from 43 frameworks, cataloging 700+ AI risks. It offers a unified tool for stakeholders to manage evolving AI risks through Causal and Domain taxonomies. This living database aims to bridge gaps in understanding and mitigating AI-related dangers. [from Massachusetts Institute of Technology & others; cc Luminos.AI Luminos.Law] ? https://lnkd.in/e2_fgXiy

AI incidents are coming. Is your organization ready? New playbooks needed for AI-specific issues. Define thresholds, monitor closely, plan containment. Consider unified AI alignment platform to manage risks across bias, privacy, safety as you scale AI. #AI #cybersecurity #LLM #GenAI Andrew Burt Luminos.AI Luminos.Law ? https://lnkd.in/ggvksQ2m