Seekr Technologies CEO Pat Condo recently announced a partnership with Bear Grylls to promote digital media literacy among young people. In a conversation with Digital, Condo shed light on the importance of understanding digital technology and its potential risks. This comes as the Biden administration takes steps to establish an artificial intelligence (AI) safety commission to address concerns surrounding this emerging technology.
While experts agree that the establishment of an AI safety commission is a necessary step, they also emphasize that it may not be sufficient to tackle all the risks associated with AI. Phil Siegel, founder of the Center for Advanced Preparedness and Threat Response Simulation (CAPTRS), suggests that the main risk lies not in the algorithms themselves but in their users’ potential misuse or malicious intent.
To address these concerns, President Biden signed an executive order that includes measures to protect Americans from the potential risks of AI systems. These measures require companies to notify the government when training new models and sharing results of “read-team safety tests.” The White House aims to ensure that AI systems are safe, secure, and trustworthy before they are made available to the public.
In addition to the executive order, the administration has announced the establishment of the AI Safety Institute, which will operate under the oversight of the National Institute of Standards and Technology. This institute will set rigorous standards for extensive red-team testing to ensure safety before the public release of AI models. Secretary of Commerce Gina Raimondo emphasizes the need for the private sector to step up and address the known and emerging risks associated with “frontier” AI models.
Experts compare the Biden administration’s approach to that of an airline checking the safety of a plane. They argue that it is not enough to only check the algorithms; procedures for users must also be evaluated. One suggestion is to require tech providers to provide user application safety procedures, similar to how banks implement “know your customer” procedures to prevent money laundering.
Critics of AI have raised concerns about its potential pitfalls, including deepfake technology’s impact on elections, the generation of child abuse material, and the use of AI-generated algorithms to breach complex digital security systems. The current system lacks a transparent appeals process and may be susceptible to political biases, as the AI safety commission’s management is under the control of the sitting president.
As the Biden administration takes action to address AI safety concerns, it remains crucial to balance technological advancements with the protection of individuals’ privacy and security. The establishment of the AI safety commission and the AI Safety Institute are steps in the right direction, but continuous evaluation and improvement will be necessary to ensure the responsible and secure use of AI.
