Digimasters Shorts - Microsoft forces AI on employees, YouTube’s AI deepfake scandal explodes, NHS uses AI for patient safety, UK buildings cyber-attacked more, OpenAI loses talent to Meta frenzy

Welcome to Digimasters Shorts, your quick hit of the latest developments from the digital world. Hosts Adam Nagus and Carly Wilson bring you insights on big tech moves like Microsoft's push to embed AI into employee evaluations, controversies surrounding AI-powered tools like Copilot, and the ongoing challenges of AI adoption among users. We cover the rise of AI-driven misinformation on platforms like YouTube, including sensational fake celebrity content and the ethical implications of automated channels. Plus, stay informed on groundbreaking health tech initiatives like the NHS’s use of AI for patient safety, increasing cybersecurity threats across UK businesses, and the fierce competition between OpenAI and Meta for top AI talent. Tune in for concise, expert updates shaping the future of technology and digital culture.

Transcript

Adam N2: 0:05

Welcome to Digimasters Shorts, we are your hosts Adam Nagus

Carly W: 0:09

and Carly Wilson delivering the latest scoop from the digital realm. Microsoft is reportedly integrating AI usage into employee performance evaluations, according to an internal memo from Julia Liuson, Corporate Vice President of the Developer Division. The memo states that using AI is now a core part of every role within the company, comparable to collaboration and data-driven thinking. This move aims to encourage adoption of AI tools among Microsoft employees, especially as these technologies have struggled to gain traction among general users. One controversial feature, Copilot's Recall, which takes periodic screenshots to track user activity, sparked security concerns and forced Microsoft to delay its release. Despite efforts to address privacy issues, skepticism remains about the tool's security. Copilot itself lags behind competitors like Chat G.P.T, boasting only 33 million users compared to Chat G.P.T’s 400 million. Windows users have expressed frustration over Copilot’s default installation in Windows 11, perceiving it as unwanted bloatware. Critics argue that Microsoft is forcing AI integration on users who are largely uninterested. This wide resistance highlights the challenge Microsoft faces in normalizing AI tools across its platforms and workforce.

Adam N2: 1:25

Dozens of YouTube channels have been generating millions of views by mixing AI-generated images and false claims about Sean"Diddy" Combs' trial, according to investigative reports. These channels produce sensational videos falsely linking celebrities like Brad Pitt, Will Smith, and Justin Bieber to the trial, often using fabricated quotes and deepfake thumbnails. Data shows 26 channels posted nearly 900 such videos, amassing around 70 million views in the past year. Many of these channels are either newly created or repurposed and monetize their content, exploiting YouTube’s ad revenue system. The practice, known as AI"slop," leverages automation tools for scriptwriting, image and video generation, and AI voiceovers, allowing low-cost mass production. Some channels, like Peeper and NV Historia, focused heavily on Diddy content but have faced demonetization or termination. While lucrative, this strategy risks legal challenges and content removal by YouTube for policy violations. YouTube has taken action by terminating at least 16 channels and demonetizing several others following investigations. Experts note that the trend reflects broader shifts toward automated, faceless channels that monetize sensationalism and misinformation. Despite the ethical concerns, operators creating this content argue that success in monetizing such videos is respected within the YouTube automation community. The NHS is set to become the first health system globally to use AI to analyze hospital data and detect potential safety issues early. This initiative is part of a 10-year plan to improve standards in mental health and maternity services following a series of recent scandals. From November, NHS trusts will launch a new"signal system" using real-time data to monitor stillbirths, neonatal deaths, and brain injuries. Health Secretary Wes Streeting emphasized that even one patient safety lapse is unacceptable and highlighted A.I's role in enabling rapid inspections and preventing harm. The move follows significant maternity care failures, including the Ockenden review's findings and a£1.6 million fine for Nottingham University Hospitals. AI technology is already used in cancer detection and quicker stroke diagnoses, signaling broader NHS digital transformation. While experts praise the AI system's potential, nursing leaders caution it should not replace essential frontline staffing. The government hopes the technology will also help reduce waiting times and facilitate the NHS's shift from analogue to digital. This development reflects a growing commitment to using data and innovation to enhance patient safety and healthcare outcomes across England.

Carly W: 4:02

More than one in four U.K businesses experienced a cyber-attack on their buildings in the past year, a significant increase from 16% the previous year, according to the Royal Institution of Chartered Surveyors. In a survey of over 8,000 business leaders, nearly 73% anticipate a cybersecurity incident will disrupt their operations within the next two years. Rics highlighted cybersecurity and digital risks as major and rapidly growing threats for building owners and occupiers. High-profile attacks, such as the one on Marks Spencer in April, forced the retailer to halt online orders for nearly seven weeks, causing a sharp drop in clothing sales. The sophistication of cybercrime, coupled with advanced technologies like artificial intelligence, is intensifying threats to critical infrastructure and data security. Worryingly, outdated operating systems, like Windows 7, are still in use in some buildings, leaving them vulnerable. Buildings have evolved into complex digital environments, gathering vast amounts of data to improve efficiency but simultaneously increasing potential entry points for hackers. Operational technologies, including building management systems, CCTV, and IoT devices, are key areas of risk, from lighting and climate control to access protocols. Paul Bagust, head of property practice at Rics, warned that failure to address these digital vulnerabilities could lead businesses to"sleepwalk" into cyber-attacks. The report calls for urgent action to strengthen cybersecurity defenses and protect businesses from growing digital threats. Open A.I is recalibrating its employee compensation packages in response to Meta's aggressive recruitment efforts. The social media giant has been offering lucrative signing bonuses and perks, successfully attracting at least eight Open A.I researchers. Among those who left is senior researcher Trapit Bansal, a key contributor to Open A.I's o1 reasoning model. Open A.I's Chief Research Officer, Mark Chen, expressed feeling like something has been stolen, highlighting the urgency of the situation. Open A.I's leadership, including C.E.O Sam Altman, is engaging directly with employees who have received Meta offers to encourage them to stay. The company is also reassessing employee performance and talent to provide better recognition and updated compensation. Meta is building a new"AI superintelligence team" led by Scale AI C.E.O Alexandr Wang, following a$15 billion investment. Altman initially believed Open A.I's mission would retain staff despite Meta's incentives but is now seeing cracks in that confidence. This competition marks a new battleground between the two tech giants focused on securing top AI talent. As the war for AI expertise intensifies, both companies are doubling down on efforts to retain and recruit key staffers.

Don: 6:48

Thank you for listening to today's AI and Tech News podcast summary... Please do leave us a comment and for additional feedback, please email us at podcast@digimasters.co.uk You can now follow us on Instagram and Threads by searching for@DigimastersShorts or Search for Digimasters on Linkedin. Be sure to tune in tomorrow and don't forget to follow or subscribe!