One of the key governance tenets around the use of AI is the concept of having a “human in the loop” to query AI-generated responses and to ensure that AI outputs remain explainable, unbiased, and harmless. But there are obvious questions: Which “human” should it be left to—and should it be one person?
Technology
Posted inTechnology
FCC bans on foreign-made robotic devices, routers, and power inverters extend compliance risks
The move confirms a trend towards expanding national security controls into broader product categories that will affect many more compliance teams than previously.
Posted inArtificial Intelligence
Deloitte survey exposes compliance gaps as a third of employees admit using shadow GenAI at work
Nearly a third of generative AI users bring their own GenAI tools to work and use them without their employer’s knowledge, raising important compliance concerns over issues such as cyber risks, copyright and audit trails.
Posted inArtificial Intelligence
California presses ahead with executive order for AI safeguards, cites federal government “abject failure”
California Governor Gavin Newsom has issued an executive order to explore whether companies should develop a “kill switch” for any potentially harmful AI tech they develop.
Posted inTechnology
Strange Polymarket trades, a KPMG employee, and a prediction market insider trading case
At a time when U.S. enforcement authorities and regulators are more closely scrutinizing insider trading in the prediction-markets space by public-company employees, another alleged insider-trading case is unfolding, this time concerning a KPMG employee.
Posted inManufacturing
Manufacturers of EU digital products must now report cyber incidents within 24 hours
If you make or sell digital products in the EU, you must now report all exploited vulnerabilities and “severe” incidents that impact their security within 24 hours of becoming aware of them. These rules apply to every sector that produces products with a digital element – from baby monitors to smart watches and apps.
Posted inBest Practices
Preparing for prediction markets regulations is a safe bet
Prediction market contracts initially gained popularity as a way for people to engage in sports betting, but have evolved to mimic financial derivative products with little institutional oversight.
Posted inData Privacy
Grindr agrees $35 million settlement for U.K. complaint over serious data breaches
The tech company behind gay dating app Grindr has agreed to make a $35.2 million settlement over allegations that it shared sensitive personal medical information—including whether people were HIV positive—with third parties.
Posted inData Privacy
Uber’s driver-blocking algorithm draws nearly $1 billion Dutch privacy penalty
The Dutch data protection authority has fined Uber $959.2 million for using automatic decision-making tech to deactivate drivers’ accounts if they received poor customer reviews.
Provided by
From Risk to Readiness: Preparing Your Workforce for AI and Data Challenges
Defensible compliance programs personalize training and offer practice opportunities. The learning metrics L&D leaders look at gauge the impact of the program, not just participation. To create a stronger program, Skillsoft developed a concise brief that explains what makes up a modern compliance program.
