palletcentral

40 Pallet C e nt ral • July -Aug u st 2 0 25 OSHA CONT. AI for writing briefs as the programs have inserted citations for cases that do not exist. Similarly, an AI program could insert faulty references to regulations or to outdated OSHA Letters of Interpretation, or simply be dead wrong. And in the safety field, being dead wrong can be deadly! Other impediments to adoption of AI include the cost for smaller employers: while software may be low cost or even free, resources are needed to program, customize, implement, and train employees on AI programs. Employers must also be encouraged to use the systems without resisting these changes. Other potential flaws in the AI universe include reliance on poor or biased data when setting up systems: "Garbage In, Garbage Out." ere may also be worker pushback if they are uncomfortable with technology or fear being pushed into less meaningful roles. Some workers may fear learning new technology, there may be language barriers, and concerns about "Big Brother" surveillance and production monitoring can cause mental stress and fatigue. On June 3, 2025, Commerce Secretary Howard Lutnick announced that the former US AI Safety Institute was being transformed into the new "pro- innovation, pro-science" US Center for AI Standards and Innovation (CAISI). e new agency will evaluate and enhance US innovation of rapidly developing AI systems, and identify vulnerabilities and threats within systems developed in the US and abroad. CAISI will now be the federal government's primary point of contact for collaborative research on securing the potential of commercial AI systems. is includes working with NIST (National Institute for Standards and Technology) organizations to develop guidelines and best practices to measure and improve the security of AI systems, and working with NIST's Information Technology Laboratory and industry to develop voluntary standards. CAISI will continue to operate within NIST as well as working with other bureaus within the Department of Commerce. e United States is not alone in embracing AI and looking at guardrails for ethical use. e European Union has adopted the EU AI Act, the first regulation on artificial intelligence. e International Labour Organization

• Cover