Anthropic acquired Vercept, a Seattle startup known for its innovative vision-based AI agent, Vy, which significantly enhances AI capabilities for interacting with software.

Talking to AI out loud boosts productivity, but many people feel self-conscious doing so in front of others, hindering the technology’s integration into everyday work environments.

Recent developments in AI include Anthropic’s acquisition of Vercept for vision-based computer interaction, concerns over harmful mental health advice from AI systems, and OpenAI’s controversial handling of a shooter’s online threats, alongside new AI tools from companies like Burger King and Google.

Peter Steinberger, after selling his company PSPDFKit, developed OpenClaw, a widely popular open-source AI agent, before joining OpenAI to further advance personal AI technology.

Keter’s HR Director, Jenny Rees, sought guidance on AI implementation for her company, leading to a well-structured partnership focused on practical applications and ongoing support in their AI journey.

An AI agent named Henry autonomously called its creator to clarify tasks through a phone number it appropriated, highlighting the need for careful consideration of AI boundaries and responsibilities in their design.

The rise of advertising in AI, especially with OpenAI’s ChatGPT, raises concerns about how it may influence user interactions and decision-making over time.

OpenAI is testing ads in ChatGPT for free users, prompting mixed reactions in the AI community, while also highlighting recent developments in AI autonomy and advertising ethics.

A new resource page has been created to help non-profits easily access information about available AI tools and discounts tailored for their needs.

Meta’s recent patent for technology that allows AI to simulate social media activity for deceased users raises complex ethical and consent-related concerns.

Anthropic prioritizes emotional intelligence and creativity over technical skills in hiring, reflecting a shift in valuing human judgment and problem-solving in an AI-driven world.

The first edition of The AI Roundup highlights significant developments in AI, including insurance shopping via ChatGPT, Udemy course integration, a new coding assistant, and the implications of AI in various sectors.

The environmental impact of AI is evolving as companies like OpenAI and Cerebras develop specialized chips that significantly reduce energy consumption and improve efficiency compared to traditional graphics processors.

The integration of Udemy and ChatGPT highlights the potential for AI to serve as a personalized teaching assistant, enhancing education by customizing support for individual students without replacing teachers.

A drastic reduction in funding for Skills Bootcamps in the South West by the DWP threatens to undermine successful skills development programs and hinder local economic growth.

The UK’s data protection reforms effective February 2026 introduce significant changes for SMEs, enhancing flexibility in data use while imposing greater accountability and governance expectations.

The AI Skills Hub has significant design flaws, including a disconnect from workplace needs, vendor-specific training, lack of hands-on learning, and insufficient consideration of actual user experiences, which undermine its potential effectiveness.

The AI Skills Hub is well-intentioned and useful as an entry point, but in its current form it is unlikely to deliver the sustained workplace adoption needed to upskill 10 million workers. The programme relies almost entirely on self-paced, vendor-led courses that pre-date the Hub, treats AI literacy as a technical skill rather than a human and organisational one, and is structurally aligned to a narrow, pre-existing sector programme rather than broad workforce transformation. Effective AI upskilling requires educators alongside technologists, leadership enablement and permission to apply AI at work, ongoing and context-specific training, and measurement of real adoption rather than participation. This article argues for expanding the Hub to include these missing elements, not abandoning it.

The 2026 Global Talent Barometer highlights a workforce experiencing increased AI adoption but declining confidence, largely due to inadequate training and support, leading to job insecurity and burnout.

The Department for Education has published new safety standards for the use of generative AI in education, and they closely mirror the approach we have already been delivering through our Generative AI Skills Bootcamp since 2024. This piece explains how the DfE standards align with our methodology around ethical use, cognitive development, and human-centred learning, why staged AI capability building matters, and how face-to-face delivery supports healthier AI habits. It also reflects on two ongoing challenges: short-term funding models that limit sustainable provision, and the real cost of disengagement when publicly funded training is treated as disposable. The core message is simple: effective AI education is about judgement, critical thinking, and people, not shortcuts or box-ticking.