Andrea Ravaioli’s Post

In order to comply with data protection regimes, AI chatbots and associated machine learning applications will have to be capable of forgetting what they've learned. Citing the "right to be forgotten" or right to erasure under Europe's GDPR, academics argue that large language models, such as OpenAI’s ChatGPT, Google’s Flan-T5, Meta’s LLaMA, and Anthropic’s Claude, and the applications integrating these models will find compliance challenging because they process and store information in a way that's different from search engines. #dataprotection #righttobeforgotten #GDPR https://rb.gy/9zdvb

The basic tenements of the GDPR remain true in the era of AI. You can’t have auto-generated logic without ingesting data. In feeding models to provide artificial intelligence what it needs to then yield accurate/innovative results, your personal data becomes even more valuable. Those companies that ignore pillars such as transparency and privacy by design do so at their own peril.

🔍 Keeping a close eye on the evolving regulatory landscape for AI. OpenAI's response to the Italian authority's notification of violation emphasizes alignment with GDPR and an intention to work constructively. However, with the possibility of fines and orders for operational changes looming, the outcome of this case may shape future AI regulatory frameworks. It's increasingly important to acknowledge #responsibleAI practices and compliance with data protection regulations. The current scrutiny of AI models, especially their data processing and legal bases for training, underscores the need for transparency, ethical considerations, and compliance within the AI community. Stay tuned for updates as the AI landscape continues to evolve 🗞️ TechCrunch: https://okt.to/trVpni

AI regulation is underway. Excellent summary of the latest from Legalweek with insightful commentary from panelist Jason Winmill, Buying Legal Council Chair.

#AI can augment human capabilities today, but it shouldn’t replace human oversight yet, especially as #AIRegulations are still evolving globally. As #GenAI becomes more widespread, there will be new and growing concerns around data privacy and ethical quandaries. https://wbmbyl.com/i3iq9X

#AI can augment human capabilities today, but it shouldn’t replace human oversight yet, especially as #AIRegulations are still evolving globally. As #GenAI becomes more widespread, there will be new and growing concerns around data privacy and ethical quandaries. https://wbmbyl.com/i3iq9X

🔍 Keeping a close eye on the evolving regulatory landscape for AI. OpenAI's response to the Italian authority's notification of violation emphasizes alignment with GDPR and an intention to work constructively. However, with the possibility of fines and orders for operational changes looming, the outcome of this case may shape future AI regulatory frameworks. It's increasingly important to acknowledge #responsibleAI practices and compliance with data protection regulations. The current scrutiny of AI models, especially their data processing and legal bases for training, underscores the need for transparency, ethical considerations, and compliance within the AI community. Stay tuned for updates as the AI landscape continues to evolve 🗞️ TechCrunch: https://okt.to/5eN0Io

We stand on the brink of a new wave of human potential, fueled by AI's power. To unlock its potential, we must learn from data privacy laws and establish global standards for ethical AI use. https://ow.ly/CpV250PrSZr #AiThority #News #informationtechnology #machinelearning

The Italian data protection agency has notified #OpenAI that it suspects ChatGPT of violating EU privacy laws following an investigation, specifically citing issues with the legal basis for processing the personal data used to train the algorithms. OpenAI has 30 days to respond. Further details in the article link below. For enterprises developing AI systems, this case highlights the importance of thoroughly vetting any training data sources to ensure legal / ethical practices that respect privacy - to meet legal compliance as well as to adhere to data governance legislation. Strict data compliance, governance checks (before issues arise) , and contingency planning if some data practices are deemed problematic should all be priorities when planning AI projects. https://lnkd.in/eDQYyNfD #ai #enterpriseai #aigovernance

We desperately need AI FOR SECURITY and SECURITY FOR AI! I recently had the chance to connect with SVB (Silicon Valley Bank), AWS Ventures, and KPMG and I’m really loving the trending investments in AI SECURITY, something only a few of us are working on behind the scenes so that we can help scale these wonderful GenAI based applications. Why only a few of us? Because the topic of GenAI Security is not sexy unfortunately, I get it! However, it’s the #1 reason why most companies are still stuck, the possibility of UNINTENTIONAL CONSEQUNCES is too great of a hurdle for them bear. It’s also the topic we CDO’s, CIO’s, and CAIO’s spend most of our time on, how do we reduce risk and exposure with: 👆🏼Data Leakage 👆🏼Human in the Loop 👆🏼Data Application & Management Layer 👆🏼API outages 👆🏼Workflow overloads 👆🏼Malicious code in Open Source code sets Not to mention all the recent regulations that just came out: 👆🏼31 countries recently passed AI regulations 👆🏼15+ are debating AI laws 👆🏼25 USA states introduced AI bills While security needs to be addressed, I hope we don’t overcomplicate it like we did with GDPR (I called it the “Gosh Darn Pain in my Rear” initiative, that consumed thousands of hours and millions of dollars for me and my teams to deploy) Signup for my Newsletter SolRashidi.Substack.com Pre-order your book 'THE AI SURVIVAL GUIDE' (https://amzn.to/3Ot2NAT). Its a HOW TO playbook for deploying PRACTICAL AI capabilities using lessons learned from REAL WORLD deployments, geared for business leaders & business owners who are not technical, and practitioners who want to learn how to make a business case.