Plan every chance, including the worst
Things happen. No matter how good and perfect Ai policy will be violations, and there will be problems. The company Chatbot will mean something embarrassed or make the promise the company can’t keep it because the appropriate guardrails are not working.
The priest says: “You hear certain exciting and wonderful examples where Ai is going well,” said the priest. “But it is a very small part of the discussion, for there are reasonable ways to treat those risks. And if there is a dose of those dangers that show, he makes countermeasures in the layouts, the training component.”
And just as a company needs to have technology in the area where AI travels on a track, AI policy also requires an application in the case of the case, and responding to employees, deliberate customers or accidental administrators.
For example, employees in certain departments may forget to review documents before they can be sent to customers, or business unit may set up the Shadow AI system that ignores data requirements or security requirements.
“Who do you call?” He asks Shellman’s Desai.
There needs to be processed, and training, to ensure that people are in a contract and power they need to support. And if there is a problem with the whole AI process, there should be a way that the system is closed without doing damage to the company.
Plan a change
Ai technology is fast. That means much of what is happening in AI’s company policy need to be reviewed and regularly updated.
“If you create a policy without the end date, you are hurting,” said Rabid Ghani, a professor in Carnegie Mellon University. That can mean that a particular provision is updated annually – or every quarter – to make sure you are still working.
“When planning a policy, you should meet the changes and need reviews,” he said. Changes can be the effect of technical development, or new business needs, or new principles.
At the end of the day, AI policy should grow new and development, can prevent it, Sectair Scoller, Principal Ey. “Anyone in the higher area – CEO or CSO – must say, ‘We will say,’ We will take AI policy to prevent you from give AI ‘,” said.
Source link