IDC released a new framework describing the skills required for employees to orchestrate AI agents in the workplace. The report notes that while many organizations are deploying agentic AI, few have formally trained staff for the new responsibilities. The framework divides the work into five areas: frame, delegate, supervise, verify, and govern. It highlights that…
A new paper from Hugging Face reviews 64 multimodal agent benchmarks and finds that only three explicitly measure whether an agent can verify its own outcomes. The study introduces the PAPAV framework, which includes Perceive, Anticipate, Plan, Act, and Verify as core capabilities. The authors note that most existing tests focus almost entirely on task…
A new article argues that the primary gap in AI adoption is managerial rather than technical. The author contends that while employees can learn to prompt systems, managers must develop the ability to evaluate the quality of generated work. Microsoft's 2025 Work Trend Index predicts that 51% of managers will train and manage AI agents…
A research team working with Apex Fusion reproduced the MIT SwarmWorld experiment using an open-weight model on four RTX 3090 GPUs. The study confirmed that AI agents in a shared environment build more resilient technology portfolios than solo agents, even with a compressed model. The replication found that the performance gap between group and individual…
New findings from HP's Work Relationship Index indicate that workers are prioritizing adaptability and critical thinking as AI adoption grows. The report notes that nearly six in ten employees now value transferable skills over traditional career progression. HP LIFE, a free digital skills program, has reached over one million learners in 2026, with high demand…
An MIT team published a study in PNAS showing how learning machinery can form from simple local connection rules. The proposed mechanism, called a Self-Assembling Motif, allows networks to organize without every component being explicitly designed. The model achieved performance similar to standard backpropagation on pattern identification tasks. Researchers caution that this computational result does…
Google announced a new industry-specific Gemini agentic artificial intelligence offering for government services, currently in preview for federal, state, and local agencies. The platform includes specialized skills, data connectors, and a partner ecosystem to automate workflows like policy interpretation and procurement support. Google Public Sector CEO Karen Dahut noted the technology is already used for…
The Department of Health and Human Services announced it will award a sole-source contract to Anthropic to maintain continuity of its Claude AI services. The agency stated that Claude is the only tool capable of meeting its specific needs for chat, drafting, and data analysis workflows. This interim measure is in place while HHS prepares…
The Department of Homeland Security issued a notice planning to prohibit nonmilitary federal agencies from using technology developed by DeepSeek. Secretary Markwayne Mullin cited risks of data collection and potential transfer of sensitive information to China. The directive was issued under the Federal Acquisition Supply Chain Security Act of 2018 and includes exceptions for the…
The Government Accountability Office released its 2027-2032 strategic plan, designating the oversight of artificial intelligence and emerging technologies as a top priority. Acting U.S. Comptroller General Orice Williams Brown stated that the agency must remain agile as federal agencies reimagine their missions using new technologies. The plan outlines goals to assess the national security implications…