Safeguarded AI (Phase 1)

The Systemic AI Safety Grants programme is a joint initiative of the UK AI Security Institute (AISI), the Engineering and Physical Sciences Research Council (EPSRC), and Innovate UK, designed to safeguard societal systems during the rapid advancement and adoption of AI technologies. Unlike AISI's Alignment Project which focuses on preventing dangerous behavior in individual AI models, the Systemic Safety programme addresses broader societal risks from widespread AI deployment across critical sectors. The programme supports research across multiple priority areas: evaluations of dangerous capabilities and safeguards, studies on user interactions with AI models, risk governance through protocols and safety cases, sector-specific AI integration risks in education, healthcare, and finance, AI agent interactions and critical infrastructure vulnerability, AI-generated misinformation, and labour market impacts. The first phase selected 20 projects and those are currently underway. With a total fund of £8.5 million, the programme's initial £4 million allocation will expand as the scheme progresses. Successful applicants receive ongoing support, computing resources where needed, and access to a community of AI and sector-specific domain experts. The programme particularly values projects bringing together academic, industry, and civil society expertise.

The ARIA Safeguarded AI programme is a flagship research initiative from the UK's Advanced Research + Invention Agency (ARIA) focused on developing machine learning capabilities with verifiable safety guarantees. The programme is structured in two phases: Technical Area 2 (TA2) on Machine Learning is currently active, with Phase 1 providing up to £1 million to up to 5 teams for initial research and development, and Phase 2 committing £18 million total for continued work on provably safe ML systems. The programme seeks to move beyond empirical testing of AI safety toward formal verification and mathematical guarantees of safe behavior. Teams must demonstrate strong organizational governance and security standards. The programme also includes TA1.4 on Sociotechnical Integration, addressing the broader challenge of ensuring safe, governable AI technology design through representative deliberation. ARIA was established by the UK government as an independent research agency modeled partly on DARPA, with a mandate to fund high-risk, high-reward research that could lead to transformative technological advances. The Safeguarded AI programme represents one of the most ambitious government-funded efforts to solve the fundamental challenge of AI safety through formal methods.

Enhancing sustainable HIV health systems and networks for quality detection, management, and monitoring to end HIV as a public health threat in India under the President's Emergency Plan for AIDS Relief (PEPFAR) is sponsored by Centers for Disease Control-GHC. Enhancing sustainable HIV health systems and networks for quality detection, management, and monitoring to end HIV as a public health threat in India under the President's Emergency Plan for AIDS Relief (PEPFAR) is a forecasted funding opportunity on Grants.gov from Centers for Disease Control-GHC. Fiscal Year: 2026. Assistance Listing Number(s): 93.067. <p><strong>The Award Ceiling for Year 1 is 0 (none). CDC anticipates an Approximate Total Fiscal Year Funding amount of $6,000,000 for Year 1, subject to the availability of funds.</strong></p><p>This NOFO will support India’s National AIDS Control Organization (NACO) in building sustainable HIV health systems, laboratory networks, and strategic information. Support aligns with national program’s Phase VI goals to eliminate HIV as a public health threat, while safeguarding American health throug...