If you have an item that we should include in this news download, or a source we should review for future items, please email me at [email protected].
Opinion: Let AI remake the whole U.S. government (oh, and save the country)
The Washington Post (March 6, 2024)
Josh Tyrangiel advocates for a complete overhaul of the U.S. government, using artificial intelligence to address the country's systemic inefficiencies & restore public trust. Tyrangiel argues that by applying AI in various government functions, similar to the success of Operation Warp Speed, the U.S. can improve service delivery, reduce bureaucratic inertia, and ultimately rejuvenate the federal government's legitimacy & effectiveness.
Faster, more personal, proactive: How generative AI is transforming digital government for citizens
CIO (March 12, 2024)
According to research cited by Avanade, a vast majority of surveyed government and public sector organizations believe citizens prefer engaging with them through new technologies like AI, which can streamline processes, provide personalized services, and improve public safety & crisis management. AI tools hold such potential in this space, as research from Accenture found that 53% of respondents find it frustrating to access public services.
Five of this year’s Pulitzer finalists are AI-powered
NiemanLab (March 11, 2024)
This year, five Pulitzer Prize finalists disclosed using AI in their journalism processes, enforced by the Pulitzer Board’s recent mandate of AI usage disclosure. As the broader industry tackles the integration of AI in journalism, other bodies, such as the Polk Awards, are similarly considering AI disclosure policies.
Exclusive: U.S. Must Move ‘Decisively’ to Avert ‘Extinction-Level’ Threat From AI, Government-Commissioned Report Says
TIME (March 11, 2024)
A State Department-commissioned report calls for stringent regulation of AI development, citing national security risks and the potential for an "extinction-level threat." However, the emphasis on severe restrictions, such as limiting AI model training power and requiring government permissions for model deployment, raises alarms about potentially hampering innovation. From our perspective, which values AI's promise for enhancing participatory democracy, the report’s alarmist claims seem overly cautious.
MIT Program Fosters Early AI Education for Future Workforce
GovTech (March 2024)
The MIT RAISE program, led by MIT Media Lab’s Cynthia Breazeal, will educate K-12 students on AI and robotics to prepare them for a future workforce increasingly reliant on these technologies. By focusing on upskilling over fear, and developing curricula that integrate AI, MIT RAISE not only seeks to demystify AI and foster early understanding, but also emphasizes responsible & ethical use.
International Women’s Day: What it takes to innovate in the age of Gen AI
AI News (March 8, 2024)
On International Women's Day, leaders in the field of generative AI highlighted its transformative impact on marketing, noting especially the need for gender-diverse teams & strategic innovation. For example, Lisa Gately of Forrester emphasized, "Gen AI brings more power to content creation, audience engagement, and personalisation." The highlighted women also underlined the importance of clean data & human insight in harnessing AI's full potential.
President Biden Calls for Ban on AI Voice Impersonations During State of the Union
Variety (March 7, 2024)
During his State of the Union address, President Biden offhandedly proposed banning AI voice impersonations as part of his wider technology agenda. His stance reflects concerns about the misuse of AI in replicating voices without consent, echoing recent negotiations in the entertainment industry about safeguarding performers' likenesses & voices. We’ll need more specifics about Biden’s proposed guardrails before determining if regulations stay narrowly tailored to misuse & malintent.
OpenAI’s GPT is a Recruiter’s Dream Tool. Tests Show There’s Racial Bias
Bloomberg (March 7, 2024)
A Bloomberg experiment revealed that OpenAI's GPT-3.5 exhibits bias in hiring practices, disproportionately favoring resumes with names associated with certain demographics over others. The analysis exposed a weakness in AI's candidate selection process – despite AI’s potential for objectivity, it can still perpetuate existing biases. While this analysis is not necessarily new, it reinforces the need for careful oversight and bias mitigation in AI applications related to governance and employment.
India reverses AI stance, requires government approval for model launches
TechCrunch (March 3, 2024)
India has changed course on AI. The government will now mandate tech companies to obtain government approval before launching new AI models, under an advisory claiming to focus on AI-induced bias, discrimination, and threats to electoral integrity. The directive, which requires compliance and status reports within 15 days, is a big departure from India's earlier laissez-faire stance on AI regulation. As scrutiny on governments’ roles in AI integration rises, we’re likely to see more governments revert to fear-based approaches.
