72% of Physicians Use AI in Medicine, Raising Patient Trust Concerns

Key Takeaways

• 72% of 1,342 physicians surveyed use AI in some capacity, according to a recent survey
• Syracuse University pre-med student warns AI note-taking tools could harm patient trust and privacy
• Experts say AI can improve diagnosis through imaging but doctors must still verify its output

Syracuse pre-med student raises concerns about AI in medicine

A Syracuse University pre-med student worries that AI note-taking tools could harm patient trust and privacy. One doctor asked to record a conversation with AI software, and the student declined. A survey of 1,342 physicians found that 72% use AI in some capacity. Experts say AI improves efficiency but doctors must still verify its output carefully.

Pre-med student says AI should assist not replace doctors

A Syracuse University pre-med student believes AI should help doctors without taking over their roles. AI tools now allow physicians to move away from manual note-taking on computers. Some doctors use AI to record patient conversations and create detailed summaries. The student stresses that AI could improve diagnosis through imaging but warns that human verification remains essential as medicine changes rapidly.

Companies reversing AI job cuts are bringing workers back

Companies that cut jobs due to AI investment are now rehiring workers, a trend called the "AI boomerang." Firms like Klarna and Ford have brought back employees after finding AI tools could not match human skills in critical thinking and creativity. Hiring managers report a growing need for people in finance, marketing, legal, and healthcare roles. Experts say the invisible work employees do remains difficult for AI to replace.

Taxing AI to protect workers could backfire, report says

A report argues that taxing AI to save jobs could hurt workers by reducing investment and innovation. Some policymakers have proposed taxes on data centers and corporate profits tied to AI growth. However, the report says labor and capital are taxed similarly in practice, and the real gap comes mainly from payroll taxes. The report warns that revenue taxes on AI could threaten competition and economic progress globally.

AI will make skilled hackers more powerful, not create new ones

A cybersecurity report says AI will not create more hackers but will make skilled attackers far more effective. In 2026, security tests showed AI models from major providers could access real computing environments. The report warns that AI helps attackers find vulnerabilities, develop exploits, and adapt malware faster with fewer specialists. Defenders are urged to automate their own security processes to keep pace with these evolving threats.

Secure AI adoption starts with strong preparation and governance

Security experts from First Distribution and The Cloud Factory held a Microsoft Copilot webinar about safe AI use. They said organisations must check data permissions and governance before enabling Copilot. The Cloud Factory built an AI Enablement Framework that guides leaders from planning to full rollout. Their Copilot Adoption Accelerator Bundle can prepare an organisation in four weeks. The approach focuses on training staff and spotting security risks early.

AI tools are helping to support elderly care and family caregivers

Nearly 23 million working Americans now care for an older adult, and AI tools are starting to help manage that burden. AI can assist with medication reminders, daily check-ins and care navigation in long-term care settings. Brookings says AI can reduce strain on family caregivers and the direct care workforce. The report stresses that human oversight remains important for agentic AI systems. Annual long-term care costs average $104,000 for in-facility care and $69,000 for in-home care.

AI helps car dealers make better pricing and appraisal decisions

A Cox Automotive survey found 82% of dealers now use AI in some form. AI is changing how appraisers value vehicles by giving them full repair history and service data. It also helps F&I managers screen applications in real time and spot possible fraud. Derek Hansen from Cox Automotive says dealers should focus on improving existing daily decisions rather than shopping for new tools. He emphasised that people still make the final call while AI provides better information.

Fortinet uses its firewalls to inspect and secure AI traffic

Fortinet announced a new AI security service that works through its existing FortiGate firewalls. The service called FortiAIGate inspects AI interactions including text images audio video and code. It applies multilayered guardrails across prompts model responses and API activity. Expanded data loss prevention can detect 35 types of personally identifiable information. The approach lets organisations deploy AI security quickly without building new infrastructure because the firewall already sits in the traffic path.

AI adoption is growing but divisions are deepening among companies

Leaders at The Information's WTF Summit said AI is dividing companies and investors. Some firms use AI to automate tasks while others use it to build new products and services. Employees also differ with some embracing AI and others remaining skeptical. The summit brought together senior women in technology and finance to discuss these challenges. The event highlighted that AI adoption is not just a technical issue but also a social and cultural one.

Experts propose Four As framework to govern AI risks

Dr. Kirti Prakash and Dr. Ian van der Linde from Anglia Ruskin University in Cambridge, England, introduced a new governance plan called the Four As. The framework includes Action, Access, Authority and Allocation to address different ethical problems with AI. It aims to keep innovation alive while protecting against severe or irreversible harm from advanced AI. The researchers argue that simply slowing AI development is not enough and each ethical issue needs a specific type of governance. The framework draws lessons from past efforts to control powerful technologies like nuclear arms and genome editing.

Sources

• As an pre-med student, I hope AI can help doctors
• As an pre-med student, I hope AI can help doctors
• What to Consider if a Company That Made AI-Related Job Cuts Wants You Back
• Taxing AI to save workers would hurt them
• AI Won’t Create More Hackers — It Will Make Skilled Attackers More Powerful
• Secure AI starts with structured enablement
• AI Is Stepping Up to Care for the Elderly
• Where AI Helps Dealers Make Smarter, More Profitable Calls
• Fortinet turns its firewall footprint into an AI security on-ramp
• Everyone’s Using AI But Divisions Are Growing: WTF Highlights
• New 'four As' framework for the governance of AI