During a November 15–21 visit to Silicon Valley, South African President Cyril Ramaphosa asked a delegation of about 300 people to identify practical AI initiatives within 90 days of returning home. He urged delegates to turn contacts “into deals,” with clear responsibilities and measurable results, and called for investment in data centres, advanced computing, digital infrastructure and research. The group includes eight cabinet ministers, startup founders, senior officials and 27 postgraduate students. Ramaphosa also emphasized developing South African technologies, businesses and intellectual property. The delegation plans discussions at Stanford and a South Africa AI Opportunity gathering; the proposed follow-up is for identifying initiatives, not a funding or application process. Ramaphosa did not disclose the visit’s total taxpayer cost.
Education and workforce preparation feature in several AI developments. The University of Memphis, led by President Bill Hardgrave, is supporting AI across its programs, including a Master of Science in Applied Artificial Intelligence designed for students without programming experience and covering practical tools and ethics. University High School is also exploring classroom uses. Separately, researchers at Carnegie Mellon University and the University of Pittsburgh School of Medicine propose simulation-based training for medical students reviewing ambient AI-generated notes. Their November 17, 2026, article notes that such tools documented more than 2.5 million encounters in one health system in a year, while medical curricula have not kept pace. Training would cover spotting errors and omissions, assessing clinical reasoning, writing notes independently and calibrating trust in AI.
Security concerns span connected devices and automated cyberattacks. Forescout estimates about 50 million UK devices are internet-exposed and warns AI can help identify equipment flaws faster; politically motivated attack claims against Britain have risen 337%. Risks include routers, security appliances and operational equipment such as refrigeration systems and hospital devices. The article notes a Pennsylvania water utility breach in 2023 involving an industrial controller with the default password “1111,” but says there is no evidence AI was used in recent refrigeration-controller attacks. Separately, the FBI says cybercriminals and state-aligned hackers use the ARTEX AI-powered penetration-testing framework to move from targeting to privileged access in under three hours. The bureau linked ARTEX to an October ConnectWise supply-chain compromise exposing Microsoft 365 customer environments, with attempted authentication against more than 2,250 downstream tenants. It urges stronger access controls, monitoring for automated activity and reporting suspected compromises to ic3.gov.
North Carolina is creating an Office of AI within the Office of the Secretary and State CIO as part of a Department of Information Technology reorganization. Chief AI Officer I-Sah Hsieh will lead efforts to help agencies and employees use AI and guide adoption. The changes also establish customer, digital experience and operations divisions, move the Enterprise Data Office into Technology Services, and consolidate more security and risk responsibility under the state chief information security officer.
Other articles examine AI’s effects on policy, engineering and creative work. An analysis of China’s AI safety rules argues that they reflect information censorship as well as state security, with strict information controls shaping AI development. For chipmakers, integrated emulation and prototyping can bring earlier visibility into RTL, software behavior, system performance and power efficiency, supporting reliable AI system-on-chips. In Hollywood, AI tools could speed visual-effects tasks such as color correction and adding or removing objects, potentially saving time and money. SoundCloud says AI brings both risks and creative opportunities, and that artist protections must keep pace. The company says it is acting against AI spam, excessive uploads and fraud that misrepresents creators or music sources, while emphasizing copyright protection and ongoing discussions with artists and partners including Reddit and Discord.
Key Takeaways
- Cyril Ramaphosa wants South Africa’s Silicon Valley delegation to identify practical AI initiatives within 90 days and turn contacts “into deals”; the delegation includes about 300 people, eight cabinet ministers and 27 postgraduate students.
- Ramaphosa called for investment in data centres, advanced computing, digital infrastructure and research, alongside South African technologies, businesses and intellectual property.
- Bill Hardgrave supports AI education at the University of Memphis, including an applied AI master’s program for students without programming experience; University High is also exploring classroom uses.
- Medical educators propose simulation training for students reviewing AI-drafted notes, after ambient AI scribes documented more than 2.5 million encounters in one health system in a year.
- Forescout estimates about 50 million UK devices are internet-exposed and warns AI can accelerate flaw discovery; it reports a 337% rise in politically motivated attack claims against Britain.
- The FBI says ARTEX can automate attacks to privileged access in under three hours and links it to an October ConnectWise compromise involving Microsoft 365 environments and attempted authentication against more than 2,250 tenants.
- North Carolina is establishing an Office of AI led by Chief AI Officer I-Sah Hsieh to guide state adoption and help agencies and employees use AI.
- An analysis says China’s AI safety approach includes information censorship as well as state security; integrated emulation and prototyping can help teams verify reliable AI SoCs earlier.
- AI could speed Hollywood visual-effects tasks such as color correction and object removal or addition, potentially reducing time and cost.
- SoundCloud says it is addressing AI spam, excessive uploads and fraud while emphasizing artist protections, copyright and creator input through discussions including with Reddit and Discord.
Ramaphosa sets 90 day goal for South Africa AI initiatives
President Cyril Ramaphosa proposed that South Africa identify practical AI initiatives within 90 days after a delegation returns from Silicon Valley. He spoke in San Jose, California, on November 15, 2026, and called for research and skills partnerships, including cooperation with universities. The visit includes planned discussions at Stanford University and a South Africa AI Opportunity gathering focused on infrastructure, investment and innovation. Professor Benjamin Rosman of the University of the Witwatersrand said South Africa should build local expertise and ensure reliable AI use, while Ramaphosa stressed privacy, ethics, security, human oversight and democratic accountability. The visit is scheduled to run through November 21, and the proposed follow up would identify initiatives rather than provide funding or an application process.
Ramaphosa urges Silicon Valley delegation to turn AI contacts into deals
President Cyril Ramaphosa told about 300 South Africans travelling with him to Silicon Valley that they should identify practical AI initiatives within 90 days of returning home. He called for clear responsibilities and measurable results, as well as investment in data centres, advanced computing, digital infrastructure and research. Ramaphosa said South Africa should develop its own technologies, businesses and intellectual property, and urged delegates to move “from contact to contract.” The delegation includes eight cabinet ministers, senior officials, startup founders and 27 postgraduate students. The presidency said the visit runs from November 15 to 21, while Ramaphosa did not disclose the total taxpayer cost.
University of Memphis and University High explore classroom AI
The University of Memphis and University High School are exploring how to use artificial intelligence in education. University President Bill Hardgrave has supported AI across university programs, including a Master of Science in Applied Artificial Intelligence for students without programming experience. The program covers practical AI tools and ethical issues. At the start of the fall semester, University High Principal Myles Beaupre met with each grade level to discuss the school’s early efforts to use AI, saying the initiative followed Hardgrave’s direction.
AI tools raise risks for connected devices as UK hacktivist claims surge
New research from cybersecurity firm Forescout warns that AI can help find flaws in connected equipment much faster, while politically motivated attack claims against Britain have risen 337%. Forescout estimates that about 50 million devices are exposed to the internet in the UK, including many routers and security appliances. The article describes risks to equipment such as supermarket refrigeration systems, factory machinery, hospital devices and data centre air conditioning. It also cites weak passwords and poor network protections, including a Pennsylvania water utility breached in 2023 through an industrial controller using the default password “1111”. The article says there is no evidence AI was used in the recent refrigeration controller attacks.
China’s AI safety rules reflect censorship as well as state security
The article argues that Western views often wrongly describe China’s AI safety approach as focused only on state security. It says information censorship is also a central priority. The government’s strict controls on information affect how AI develops in China. The article says this broader mix of priorities has important implications for the country’s AI development.
Integrated verification helps teams build reliable AI SoCs
A continuous hardware verification strategy can help teams build AI system-on-chips, or SoCs. Integrated emulation and prototyping give chip teams an earlier view of RTL, software behavior, system performance and power efficiency. The article says a strong verification strategy helps ensure AI SoCs are high quality and reliable.
AI tools could speed up visual effects work in Hollywood
AI tools may speed up visual effects, one of Hollywood’s most time-consuming jobs. They could automate tasks such as color correction, removing objects and adding objects, potentially saving filmmakers time and money. New York Times culture reporter Matt Stevens talks with visual effects artists and a filmmaker about how AI is changing their work. The video is part of a series about AI in entertainment.
FBI warns hackers are using AI tool ARTEX for global cyberattacks
The FBI said cybercriminals and state-aligned hackers are using ARTEX, an AI-powered penetration testing framework, to automate attacks against organizations worldwide. The bureau said ARTEX can move from initial targeting to privileged access in under three hours. It linked the tool to an October ConnectWise supply-chain compromise that exposed Microsoft 365 customer environments and involved attempted authentication against more than 2,250 downstream tenants. Since at least August 2026, malicious actors have used ARTEX in sectors including finance, government, technology, defense, semiconductors, energy, health care, manufacturing, telecommunications and law. The FBI urged organizations to strengthen access controls, watch for suspicious automated activity and report possible compromises to ic3.gov.
Researchers outline training for medical students who review AI notes
Researchers from Carnegie Mellon University and the University of Pittsburgh School of Medicine proposed a framework to train medical students to review notes drafted by ambient AI scribes. Published November 17, 2026, the article says these tools documented more than 2.5 million clinical encounters in one health system in a year, while medical curricula have not kept pace with their use. The framework treats reviewing AI-generated notes as a skill that can be taught through simulation, with learning goals and assessments. It covers finding errors and omissions, checking whether notes reflect clinical reasoning, writing notes independently and adjusting trust in AI across situations. Simulation can let trainees practice identifying mistakes without putting patients at risk.
North Carolina reshapes IT department and creates a dedicated AI office
North Carolina is reorganizing its Department of Information Technology and creating an Office of AI within the Office of the Secretary and State CIO. Chief AI Officer I-Sah Hsieh will lead the office, which will help state employees and agencies learn to use AI and guide its adoption. The changes also create Customer, Digital Experience and Operations divisions, led by Torre Jessup, Lenae Storey and Keith Briggs. NCDIT is moving its Enterprise Data Office into the Technology Services Division, led by Greg Flynn, and placing more security and risk responsibility under the state chief information security officer. General Counsel Julie Song will lead a new legal division, while Pryor Gibson and Annette Taylor will continue work on broadband and community partnerships.
SoundCloud CEO Says AI Must Protect Artists and Support Creativity
SoundCloud CEO says the company wants music’s future to put artists first as artificial intelligence changes how music is made. SoundCloud sees AI as both a risk to artists and a chance to expand creativity, and says protections must grow alongside new technology. The company says it listens to artists through roundtables and conversations with creators and partners, including Reddit and Discord. SoundCloud is taking action against AI spam and excessive uploads that hurt music discovery, as well as AI fraud that misleads users about a creator or the source of music. It also says copyright protection is essential to safeguarding creators and rights holders.
Sources
- Ramaphosa proposes 90-day plan for AI research and investment initiatives
- Ramaphosa's AI investment pitch in Silicon Valley
- University of Memphis, University High School explore AI implementation
- AI makes connected devices easier to target as UK hacktivist threats surge
- What the west gets wrong about AI safety in China
- Building AI SoCs with a continuous hardware verification strategy
- Video: Will A.I. Make This Tedious Hollywood Job Easier?
- FBI warns hackers using AI-powered tool to automate cyberattacks worldwide
- As use of AI scribes spreads, researchers propose a framework for preparing medical trainees
- North Carolina Reorganizes IT Leadership, Adds Dedicated AI Office
- SoundCloud CEO on How AI Is Changing & The Future Builds on Artists
Comments
Please log in to post a comment.