Aaron Levie, the co-founder and CEO of Box, has warned that corporate leaders are making flawed decisions about artificial intelligence replacing human workers. He described a phenomenon he calls 'AI psychosis,' where executives who lack deep knowledge of specific roles assume AI can perform those jobs. Levie made the remarks during a recent interview, pointing to a growing disconnect between management's enthusiasm for automation and the actual complexity of many positions.

Levie argued that the people deciding AI can replace a job are often the least qualified to understand what that job truly entails. He suggested that this mindset leads to premature or misguided layoffs, as companies rush to cut costs by deploying AI tools without fully assessing their limitations. The Box CEO's comments come amid a wave of AI-driven workforce reductions across the technology sector.

ClickUp, a project management software company, recently announced it would cut 22% of its workforce to make room for AI agents. The company stated that the layoffs were part of a strategic shift toward automation, with AI handling tasks previously done by human employees. ClickUp's CEO Zeb Evans said the move was necessary to stay competitive, though critics argue it reflects the very 'psychosis' Levie described.

Data from layoff tracking platforms shows that tech industry job cuts in 2026 are already approaching the total number recorded for the entirety of 2025. Analysts attribute this acceleration to companies aggressively adopting AI technologies, often without clear evidence that the replacements will maintain or improve productivity. The trend has sparked debate about whether executives are overestimating AI's capabilities.

Levie's warning echoes concerns raised by other tech leaders, including OpenAI's Sam Altman and Microsoft's Satya Nadella, who have cautioned against over-reliance on AI. However, Levie's framing of 'AI psychosis' specifically targets decision-makers who lack technical or operational familiarity with the roles they eliminate. He urged companies to conduct thorough assessments before replacing human workers with AI systems.

ClickUp's layoffs are part of a broader pattern in the software industry, where companies are restructuring around AI agents that can automate customer support, data entry, and even some coding tasks. The company said affected employees would receive severance and support, but the move has drawn criticism from labor advocates who argue that AI cannot replicate human judgment in complex workflows.

Levie did not name specific companies beyond ClickUp, but he noted that many organizations are making similar mistakes. He recommended that executives spend time understanding the day-to-day realities of roles they plan to automate, rather than relying on abstract efficiency metrics. 'You can't replace what you don't understand,' Levie said.

As of early 2026, the tech industry has seen over 150,000 layoffs, with AI cited as a primary factor in roughly half of those cases. The trend shows no signs of slowing, with several major firms announcing further cuts tied to AI adoption. Levie's comments serve as a counterpoint to the prevailing narrative that AI will inevitably replace large swaths of the workforce, urging a more measured approach.

Samsung Display has introduced a new QD-OLED panel that combines 4K resolution with a 360Hz refresh rate, a first for the technology. The panel is designed for use in premium monitors, particularly aimed at gamers and professionals requiring high refresh rates and sharp image quality. The announcement was made on March 12, 2025, through an official press release.

The new panel leverages Samsung's Quantum Dot OLED technology, which uses quantum dots to enhance color accuracy and brightness. It supports a 4K resolution of 3840 x 2160 pixels and can achieve a refresh rate of up to 360Hz, reducing motion blur and improving responsiveness. The panel also covers 99% of the DCI-P3 color gamut and has a peak brightness of 1,000 nits.

Samsung Display stated that the panel incorporates advanced OLED pixel technology to deliver deep blacks and high contrast ratios. The 360Hz refresh rate is achieved through improved driving circuitry and faster response times, which are critical for competitive gaming. The panel also supports Adaptive-Sync technologies like AMD FreeSync and NVIDIA G-Sync to eliminate screen tearing.

The company plans to begin mass production of the 27-inch and 32-inch versions of the panel in the second quarter of 2025. These panels will be supplied to monitor manufacturers, including Samsung Electronics, Dell, and ASUS, for their upcoming gaming monitors. Pricing for the final products has not been disclosed, but they are expected to be positioned in the premium segment.

Samsung Display highlighted that the new panel is part of its strategy to lead the high-end monitor market. The company has been expanding its QD-OLED lineup, which already includes 34-inch and 49-inch ultrawide panels. The 4K 360Hz variant is expected to compete with LG Display's OLED panels and Mini-LED alternatives.

Industry analysts note that the combination of 4K resolution and 360Hz refresh rate addresses a growing demand for high-performance displays among esports enthusiasts and content creators. The panel's fast response time and wide color gamut make it suitable for both gaming and professional color-critical work.

Samsung Display will showcase the new panel at the upcoming Display Week 2025 event in May. The company expects the first monitors featuring the panel to hit the market in the third quarter of 2025. Specific product announcements from partner brands are anticipated in the coming months.

"This panel represents a significant milestone in display technology, offering both stunning detail and ultra-smooth motion," a Samsung Display representative said in the press release. The company aims to ship over 1 million QD-OLED panels in 2025, driven by demand for high-refresh-rate monitors.

The U.S. Centers for Disease Control and Prevention published a detailed analysis of last year's major measles outbreak that originated in West Texas, providing fresh evidence of the virus's severity. The study, released yesterday in the Morbidity and Mortality Weekly Report, examined cases from the multi-state outbreak that spread rapidly through undervaccinated communities. Researchers found that approximately one in five infected individuals required hospital care, with young children bearing the brunt of severe complications.

Health Secretary Robert F. Kennedy Jr., a prominent anti-vaccine advocate, has repeatedly dismissed measles as "just a rash" and alleged that outbreaks are fabricated to create fear. In 2021, Kennedy wrote that "Measles outbreaks have been fabricated to create fear" when the U.S. recorded only 49 cases nationwide. The new data directly contradicts those assertions, showing a disease that frequently leads to pneumonia, encephalitis, and other serious outcomes.

The outbreak emerged in an area with declining vaccination rates, a trend that has accelerated over the past two decades. The U.S. declared measles eliminated in 2000 following a highly successful vaccination campaign, but slipping immunization coverage has allowed the virus to regain a foothold. The West Texas cluster quickly spread to multiple states, overwhelming local health departments and straining hospital resources.

Hospitalization rates in the outbreak reached 20%, with children under five years old comprising the majority of severe cases. Measles complications often require intensive care, including respiratory support and treatment for dehydration. The virus is highly contagious, with each infected person capable of spreading it to 12 to 18 others in a susceptible population.

Public health officials have long warned that falling vaccination rates would lead to the resurgence of preventable diseases. The current outbreak follows a pattern seen in other countries where measles reemerged after immunization coverage dropped below the 95% threshold needed for herd immunity. The CDC report underscores the fragility of the elimination status achieved in 2000.

Kennedy's appointment as Health Secretary has raised concerns among medical professionals about the administration's commitment to evidence-based public health policy. His past statements questioning vaccine safety and efficacy have been widely debunked by scientific research. The new outbreak data provides a stark real-world example of the consequences of vaccine skepticism.

The study's authors emphasized that measles remains a serious threat, particularly for unvaccinated individuals. They urged clinicians to maintain high suspicion for measles in patients with fever and rash, especially those with travel to outbreak areas or known exposure. The report also highlighted the importance of maintaining robust vaccination programs to prevent future outbreaks.

As of the publication date, the outbreak has subsided but sporadic cases continue to be reported. The CDC continues to monitor the situation and recommends that all eligible individuals receive the measles-mumps-rubella vaccine. The full MMWR report is available on the CDC website.

The rapid advancement of artificial intelligence has introduced a flood of specialized vocabulary. Many people encounter terms like "large language model" or "diffusion" without a clear understanding of their meaning. A new glossary aims to bridge that gap by providing straightforward definitions for the most commonly used AI terminology.

Machine learning, a subset of AI, refers to systems that improve performance through experience rather than explicit programming. Algorithms analyze data to identify patterns and make decisions with minimal human intervention. Deep learning, a more advanced branch, uses multi-layered neural networks to process complex information such as images or speech.

Neural networks are computing systems inspired by the human brain. They consist of interconnected nodes, or neurons, that process data in layers. Training a neural network involves adjusting the connections between neurons to reduce errors in its output. This process requires vast amounts of data and computational power.

Generative AI describes models that create new content, including text, images, music, or video. These systems learn patterns from training data and then generate original outputs that mimic those patterns. Examples include language models that write essays and image generators that produce artwork from text prompts.

Large language models, or LLMs, are a type of generative AI trained on enormous text datasets. They can understand and produce human-like text, answer questions, summarize documents, and even write code. Models like GPT-4 and Claude are prominent examples of LLMs currently in use.

Diffusion models are a class of generative models that create images by gradually adding and then removing noise from data. Starting from random noise, the model refines the image step by step until it matches a given text description. This technique powers many popular AI image generators.

Reinforcement learning involves training an AI agent through trial and error. The agent takes actions in an environment and receives rewards or penalties based on outcomes. Over time, it learns to maximize rewards, making it effective for tasks like game playing or robotics.

The glossary also covers terms like "transformer," a neural network architecture that underpins many modern LLMs. Transformers process entire sequences of data simultaneously, enabling efficient handling of long-range dependencies in text. This architecture has become foundational in natural language processing.

Understanding these terms helps demystify how AI systems operate and what they can achieve. As AI continues to integrate into daily life, familiarity with its vocabulary becomes increasingly valuable for informed discussions and decisions.