RPC’s Post

Generative AI tools such as ChatGPT pose risks to data protection – firms still struggling to put an AI strategy in place will struggle down the line #ai #dataprotection https://lnkd.in/evPspzw7

The EU AI Act, approved by members of parliament in March 2024, defines general purpose AI (GPAI), comprising foundation models such as ChatGPT and DALL-E, the workhorses behind generative AI innovation.   To abide, and starting the middle of 2025, creators of such GPAI technology must maintain detailed documentation on testing and training processes; pass on information and documentation to downstream providers; and establish a policy to respect IP and copyright. If the AI has systemic risks, providers must also perform intensive model evaluations; track, document, and report serious incidents; and ensure adequate cybersecurity protection.   To help in this, we advise a ten-pronged enterprise response strategy to the act. Of note here is to get ahead by building responsible AI into AI systems from the get-go. This includes built-in measures for increased fairness, transparency, accountability, and regulatory compliance at all stages of AI design and development. Harry Keir Hughes Syed Q Ahmed Mona Dash Kate Bevan Ritarshi Chakraborty Infosys Topaz #AI #enterpriseAI #EUAIact #AIact #AIregulation infy.com/4ajWyYa

More data does not always give better AI results. In this article, my colleague Steve Bradley discusses the importance of cleansing and using the right data (vs. dark data) to train AI systems. Thank you Steve for shining a light on this topic.

What are the risks of ignoring the mantra "IA before AI"? One of the dangers is dark data – present in every enterprise – the redundant, obsolete and trivial information you accumulate but never use. While the data itself may be mundane, it is far from harmless, especially when you start experimenting with AI tools. In this article, Steve Bradley of Harbor explains why it's critical to bring your dark data into the light. The risks go beyond inadvertently exposing sensitive material. Your AI implementation may be sabotaged from the start if you feed the AI model biased or unreliable information. Read more about what you need to know about dark data – and how to ensure it won't become a problem for your AI initiatives: https://bit.ly/3xnvwSf (And kudos to Alex Smith of iManage for raising awareness within the legal ecosystem by coining the #IAbeforeAI hashtag.) #informationgovernance #legalops #legaltech #darkdata

Generative AI And Data Protection: What Are The Biggest Risks For Employers? via InnovationWarrior.Com #ai #enterprise_tech #innovation #AI #Enterprise_Tech #Innovation #standard #Technology

Generative AI And Data Protection: What Are The Biggest Risks For Employers? via InnovationWarrior.Com #ai #enterprise_tech #innovation #AI #Enterprise_Tech #Innovation #standard #Technology

The Generative AI Acceptable Usage Policy needs the following : a. Consumers will want to know how their information is being managed by AI-driven systems. Regulators may ask if you are adhering to principles of responsible AI. b. • What are your managers, employees, and IT department’s generative AI responsibilities? c. Protect AI systems against unauthorized access, tampering, and misuse. d. It is not the employee’s decision to share company data with generative AI tools like ChatGPT. e. If an organization is committed to providing a safe and secure environment, then any AI used must be shown to be responsible to avoid causing any type of harm. f. What are the AI ethics and principles your organization must adhere to?. g. Highlighting expectations and acceptable behavior versus unacceptable behavior is an important consideration for creating a policy. h. It’s crucial to specify guidelines when sourcing data, particularly when dealing with personal data or sensitive information (e.g., a strong emphasis on the use of de-identified and anonymized data to ensure privacy).  https://lnkd.in/gdYYKG25

MineOS launches AI Asset Discovery and Risk Assessment module to help enterprises govern AI development and usage, addressing the 'black box' problem and emerging regulations. https://lnkd.in/e3DvhjDe #AI #Governance #Enterprise #MineOS 💻💡

Business leaders are nervous about generative AI. And they're right to be. GenAI has impacts on: -Information security -Data protection -Business ethics -Legal compliance -Customer experience -Brand equity But you can manage these risks if you implement AI tools ethically and strategically. Get a list of executive action items for generative AI to make sure you get all the business boons without the setbacks in this article from Bigtincan President Patrick Welch. #generativeai #chatgpt #airisks

#Blog | As AI rapidly transforms the business ecosystem in 2024, ensuring its secure and ethical implementation has become crucial. Governments worldwide are rolling out AI regulations to address concerns like bias, data privacy, and security. From the EU's AI Act to the US's executive orders, these regulations aim to foster safe and fair AI usage. As a business owner, you must stay ahead by understanding these regulations and integrating AI responsibly. Read our latest blog to discover the key ethical considerations for AI development and how they can transform your business operations. Stay informed and competitive in the evolving AI landscape. https://lnkd.in/gRw23Dir #AI #AIRegulations #artificialintelligence #AIdevelopment #testingxperts #testingservices