"Gaul's insights into what venture capitalists want to see when investing in AI startups. "They want to know you have a maturity level with respect to your AI risk understanding, your compliance policies, so you're not going to create substantial problems down the road." From advising investors Gaul has learned investors "want some kind of assurance that executives know what they're doing and aren't going to create risk"."
"Why focusing on a proprietary single large language model (LLM) may be too expensive or limiting. Gaul believes "we're starting to see a transition there. This is a projection that I've been making that as AI continues to evolve, we're going to see people drifting to these targeted use models that have been trained on highly specific data sets." This transition is because targeted or small models are "more cost effective for companies to build on that those from businesses like ChatGPT and the Anthropics.""
"Gaul has recognized a recent shift in IP protection for small AI-driven companies: "I'm seeing people leverage more trade secrets rather than going for patenting because a lot of the magic is in your data set and that's not individually protectable.""
Venture capitalists and investors seek evidence of AI risk maturity, compliance policies, and executive assurance to avoid substantial downstream problems. Publicly facing policies for responsible AI use and compliance enhance credibility and investor confidence. Single proprietary large language models can be costly and limiting, driving a shift toward targeted, smaller models trained on specific datasets that are more cost effective for many companies. Small AI-driven companies increasingly favor trade secrets over patents because competitive advantage often resides in curated data sets that are not individually patentable. Proactive governance supports sustainable innovation and responsible IP use.
Read at IPWatchdog.com | Patents & Intellectual Property Law
Unable to calculate read time
Collection
[
|
...
]