The Data Governance Challenges of LLMs

Get ready to explore the hidden side of large language learning models! In this episode, we dive into the fascinating world of cutting-edge AI technologies and the data governance challenges of LLMs. Join us as we unravel the complexities of managing vast amounts of data, ensuring privacy, and addressing ethical considerations in the realm of large language learning models.
Our guest, Bob Seiner, will share insights and advice to navigate these challenges successfully. If you're curious about the future of AI and data governance, this episode is a must-listen. Tune in now and stay ahead of the curve!
#datagovernance #LLM
☕ If you found this content useful and would like to support the channel, you can buy me a coffee: www.buymeacoff...
**********************************
My mission at LightsOnData is to:
1️⃣ Provide current and future data professionals with increased knowledge and better understanding of data governance, data management, and data storytelling.
2️⃣ Provide simple, fun, and engaging, as well as thorough explanations, teachings, and advice on important, but otherwise boring and dry topics of the data world.
3️⃣ Put the lights on data-focused organizations and products by helping them get more exposure on social media.
Here are ways that you can benefit from LightsOnData's content:
💡 ONLINE COURSES: Easy to learn, engaging and highly valued online courses in areas of data governance, data management, and much more:
www.lightsonda...
💡 LIGHTS ON DATA SHOW: Each episode puts the lights on various data topics with renowned industry experts. Listen and watch the live show and podcast:
anchor.fm/ligh...
www.lightsonda...
💡 FREE TEMPLATES: Check out our website for more useful content, templates, best practices, definitions, etc. on data governance, data management, data storytelling practices.
www.lightsonda...
💡 DAILY TIPS: Follow our LinkedIn page to benefit from daily content on data governance, data management, data quality, data storytelling, and all things data.
/ lightsondata
💡 FUN DATA SWAG: Show others your love for data with any of these fun apparel items and accessories:
www.lightsonda...
**********************************
Connect with George Firican on Social Media:
/ georgefirican
/ georgefirican
/ lightsondata
/ georgefirican
www.pinterest....
Subscribe to LightsOnData KZbin Channel Here: / lightsondata

[@HeatherFaraMS]

Writing AI policy for government presently and there is a true gap in understanding when it comes to leadership’s understanding of the future already here with custom AI model building. I included language about custom model development, validation, and documentation that was all removed. Proprietary LLM and generative AI platforms are dropping in software with little recognition. These platforms allow people to develop custom models and train things like data classification without data quality parameters, validation, documentation, or OTB model monitoring. Developers of these models get squirmy and say things like: “It’s just not the same.” “We don’t need metadata anymore. “The model knows in ways that can’t be documented.” When the model is making determinations, there is logic that we allow to be unexposed. Governing is difficult when we collectively accept black boxes. I have proposed a human in the loop but l can’t say where in the loop that is most critical or dangerous. LLM are prone to tunnel vision. When you don’t see counter arguments and automated summaries are your background research, l am concerned that expertise and novel thought will go down over time, the general watering down of knowledge.

[@bigyoyo2030]

High level of blablabla work. Spend first 5 minutes just to agree that there is no almost difference. Good job guys.