42% of executives in a new survey say that AI adoption is tearing companies apart. Cursors max mode unlocks the full power of code 3.7 for AI assisted coding, but at a cost. And Jensen Huang Unleashed. He says, AI compute needs are underestimated and he eyes quantum computing's Future. Welcome to hashtag Trending. I'm your host, Jim Love. Let's get into it.
A new survey by writer, AI and Workplace Intelligence reveals that while enterprises are eager to integrate generative ai, the adoption process is creating some deep divisions with 42% of executives admitting it's tearing their companies apart, Despite widespread optimism with 88% of employees and 97% of executives saying they've benefited from ai, many companies struggle with internal misalignment, resistance, and a lack of strategic direction.
The survey conducted in December, 2024 gathered insights from 1600 knowledge workers across US enterprises. While enthusiasm for AI remains high power struggles and poor execution are hindering success. Nearly two thirds or 63% of executives say AI adoption has created tension or division within their company, and 71% admit AI tools are being built in silos. Limiting their effectiveness. Employee resistance is also a major issue.
41% of Millennial and Gen Z workers are actively sabotaging their company's AI efforts. According to the study, citing concerns about job security, poor AI quality, and lack of trust in the technology, some are refusing to use AI generated outputs while others manipulate results to make AI appear less effective. Security is another challenge. 35% of employees report using non-approved AI tools, often paying for them out of pocket, which raises compliance and cybersecurity risks.
Meanwhile, 73% of companies are investing over 1 million annually in ai, but only one third report seeing significant ROI. The survey underscores a clear disconnect between leadership's vision and real world implementation. Successful AI adoption. The report suggests depends on clear strategy, better communication, and high quality AI tools that genuinely support employees rather than disrupt workflows.
AI powered code editor cursor from Anthropic has introduced max mode for Anthropics code 3.7 sonnet, enabling developers to process larger code bases with a 200,000 token context window and an expanded tool calls While this upgrade enhances AI assisted coding, it comes with a usage based pricing model that could significantly impact costs. Cursor is an AI powered code editor built on Visual Studio Code or VS.
Code, designed to help developers write debug and refactor code with AI assistance, It integrates models like code from anthropic or GPT from open AI to provide smart code completion, debugging and AI powered edits. Cursor's deep integration with VS. Code makes it a seamless upgrade for developers looking to enhance productivity without leaving their preferred IDE.
Cursor's New Max mode leverages code 3.7, expanded context window, and increased tool call limit to implement AI powered coding assistance with 200,000 tokens of memory code 3.7 can process entire projects making it ideal for large scale development and complex debugging. Additionally, the tool call limit has been raised to 200 per chain, allowing AI to make multiple changes in one session. However, Max Mode is not included in Cursor Standard Pro Plan.
Instead, users must enable usage-based billing with each prompt costing 5 cents plus 5 cents per tool call. Now with 200 tool sets available, developers working on extensive projects could see these charges accumulate very quickly. Despite the pricing concerns cursor sees Max Mode is a critical upgrade for developers handling large repositories or those who need advanced AI reasoning for code refactoring and optimization.
The move aligns with a broader trend in AI driven development where models like Code 3.7 are improving code comprehension reasoning, and long form context Cursors rollout of max mode highlights both the promise and the challenges of AI assisted coding. It offers unprecedented power, and it's not open AI's $20,000 per month for a PhD level resource, but it's still at an increased price that developers will need to weigh carefully. NVIDIA's, CEO Jensen.
Huang said that he was going off teleprompter and he meant it He delivered a speech at NVIDIA's GTC conference that was more frank and perhaps more controversial than we might be used to. Huang said predictably that he believes the world is vastly underestimating the computer power needed for ai, But the shock from this announcement was quantum computing as a critical part of the future. Only a few months ago, Huang was saying that quantum computing was 15 to 30 years out.
To our surprise, Huang emphasized the company's growing interest in quantum computing, hinting that AI's future may require breakthroughs beyond traditional processors. And Huang also reinforced his longstanding argument that demand for AI compute will accelerate, in that light, he unveiled NVIDIA's Blackwell architecture, which is designed to significantly boost AI performance and efficiency for large scale models.
he said that Nvidia also plans to release a new chip set called Rubin, GPU in 2026. The chip is named after Vera Rubin. The American astronomer known for her research on dark matter, and Huang referred to himself as the revenue destroyer. Claiming half jokingly that when this new chip architecture hits the market, Nvidia won't be able to give away the previous generation hopper chips.
Not exactly the speech that salespeople want to hear, and following There will be yet another upgrade named after Richard Feynman, the renowned physicist whose integral to the theories of quantum physics. So Huang's comments and quantum computing stood out. While, NVIDIA remains a dominant force in classical computing with GPUs. Huang acknowledged that the future AI models may require hybrid architectures that combine quantum and traditional computing.
This aligns with growing industry speculation that quantum computing could help solve AI's increased computational bottlenecks. Nvidia has already begun experimenting with quantum simulation techniques using GPUs. While Huang stopped short of announcing a dedicated quantum division, he hinted that NVIDIA is actively researching quantum acceleration to push AI forward.
this could position Nvidia as a major player in quantum ai competing with IBM, Google and other firms racing to commercialize quantum technology. Despite concerns that NVIDIA's rapid advancements in GPUs would outpace market demand, Huang remains confident The world is not overestimating. AI compute needs. It is underestimating them. He said, and his comments suggest that Nvidia sees no slowdown in AI infrastructure spending and is already preparing for the next wave of computing evolution.
and that's our show for today. Thanks for joining us. Always happy to hear from you. You can reach me at [email protected]. I'm your host, Jim Love. Have a wonderful Wednesday.