The popular press continues to reverberate with stories about the miracles of generative artificial intelligence (GAI) and machine learning (ML), and all the ways it might be used for good—and for bad. There’s hardly a tech company that isn’t talking about how GAI/ML can enhance its offerings. And those same companies, from the board to marketers and developers, are wrestling with how to leverage this new technology without facing an avalanche of unintended consequences.
Contemplating all those same questions at Synopsys, we rounded up a faculty of technology lawyers and technologists (some in the same body) to explore this dynamic space. We quickly realized there was just too much to talk about in one webinar, so we made it a four-session online course (and the syllabus was still overflowing).
AI strategy, security, and governance: The view from the top
The course focuses on how to manage the risks of using GAI to develop software, but it starts high level and broad. The first session defines the space and the issues, and then examines how to educate boardroom and C-suite types about how and why they should be thinking about the issues. In sum: Don’t ignore; don’t freak out. A big takeaway that resonated throughout was: Humans can already make trouble in the same ways that an AI can, so companies can build upon the human-directed governance that’s already in place.
Best practices for using AI in software development
The second session builds on the how ML and GAI work and how GAI can assist in writing software. It then covers the range of legal issues associated with incorporating machine-written code in a company’s software and surveys the numerous GAI-related cases currently in the courts and their implications. It concludes with recommended best practices in light of the uncertainty and risks.
GAI, training data, open source, and GitHub Copilot, oh my!
The third session explores the nuances and licenses for models and training data—key considerations for companies building AI into their applications or using it for coding. It then shifts to a deep dive on the Copilot class action suit, perhaps the most consequential case for companies looking using AI in development.
Ask the experts: AI and software development
The final session of the course brings the faculty back together for a panel discussion of the most interesting questions to come out of the first three sessions of the course.
Continue Reading
Audited vs. automated: What your automated open source tool isn't seeing
The hidden business risks of technical debt in mergers and acquisitions
From diligence to integration: How software audits inform post-close M&A strategies