• AI KATANA
  • Posts
  • AI slop is suffocating the web, says a new study

AI slop is suffocating the web, says a new study

Also: How to hedge against AI stealing your job

Author

AI KATANA
March 10, 2025

Here’s Why Over 4 Million Professionals Read Morning Brew

  • Business news explained in plain English

  • Straight facts, zero fluff, & plenty of puns

  • 100% free

Hello!

The internet is being overwhelmed by AI-generated content, according to a new study that highlights concerns about declining information quality and creativity. Meanwhile, Foxconn has unveiled a traditional Chinese large language model, FoxBrain, to enhance AI-driven manufacturing, and Apple has delayed key Siri improvements in iOS 18 due to infrastructure challenges. As AI continues reshaping industries, a new report explores strategies for workers to safeguard their careers against automation, while China is ramping up university enrollments in AI and other strategic fields to strengthen its technological capabilities. In real estate, an AI agent has facilitated $100 million in property transactions, showcasing the technology’s growing role in high-value sales. Elsewhere, scientists have developed an AI-powered digital twin that can control and adapt to its physical counterpart, marking a major leap in autonomous systems. On the venture capital front, ServiceNow is set to acquire Moveworks for $2.85 billion to enhance enterprise AI capabilities, while DeepSeek remains independent, avoiding VC funding despite industry interest.

Sliced just for you:

  • 🌐 AI slop is suffocating the web, says a new study

  • 🏭 Foxconn launches traditional Chinese large language model for AI-driven manufacturing

  • 🍏 What went wrong with Apple Intelligence Siri development?

  • 🛡️ How to hedge against AI stealing your job

  • 🎓 China’s top universities expand enrollment to beef up capabilities in AI, strategic areas

  • 🏡 How an AI agent reached $100M in property deals

A new study warns that the internet is being flooded with AI-generated content, raising concerns about its impact on information quality and creativity. Researchers analyzed over 300 million documents, including consumer complaints, press releases, and job postings, and found a dramatic increase in AI involvement since the launch of ChatGPT in late 2022. The proportion of AI-generated content in some domains surged from 1.5% to over 25%, with certain regions in the U.S. using AI in up to one in four complaints. The study, published on Cornell’s arXiv server, employed a custom statistical framework to detect AI-generated text with high accuracy, revealing the widespread integration of AI tools across multiple industries. Experts fear this influx of AI-generated content—often repetitive and lacking originality—could dilute online discourse and hinder human creativity. As AI-generated material becomes harder to distinguish from human-authored content, the study highlights the urgent need for transparency, content verification, and policies to manage the growing influence of AI on the web.

Foxconn has launched FoxBrain, a traditional Chinese large language model designed to enhance AI-driven manufacturing. Trained on 120 Nvidia H100 GPUs, the model specializes in mathematics and logical reasoning, aiming to optimize efficiency in industrial applications. Initially developed for internal use, Foxconn plans to open-source FoxBrain to collaborate with technology partners and expand its impact. The model is based on Meta’s Llama 3.1 architecture with 70 billion parameters and reportedly outperforms Llama-3-Taiwan-70B in traditional Chinese language comprehension benchmarks. Foxconn’s AI research team prioritized refining the training process over maximizing computing power, ensuring a more efficient approach. This move aligns with the company’s broader AI strategy, which includes leveraging Nvidia’s Omniverse for digital twin technology in factory operations and supply chain management.

Apple has officially delayed the release of key Siri enhancements in iOS 18, citing the need for additional development time without providing specific reasons. Reports suggest that the delay stems from Siri’s outdated infrastructure, which currently operates on two separate systems—one for legacy commands and another for more advanced AI-driven queries. Apple’s engineers have struggled to integrate these systems smoothly, leading to significant internal bugs and performance inconsistencies. Executives, including software chief Craig Federighi, expressed concerns that the new features were not working as advertised, which contributed to the postponement. Some insiders argue that Apple’s AI division requires stronger leadership, as the company risks falling behind competitors in AI-driven personal assistants. Instead of rushing the flawed system, Apple is now expected to integrate these capabilities into a unified Siri backend with iOS 19, likely delaying full functionality until at least late 2025.

As AI rapidly advances, knowledge workers face an increasing risk of job displacement, prompting a need for strategic adaptation. The rise of AI-powered tools is reshaping white-collar professions, automating tasks once thought to require human expertise. Sectors such as finance, law, software development, and consulting are particularly vulnerable, with companies already reducing hiring and implementing AI-driven efficiencies. To remain relevant, professionals must integrate AI into their workflow, enhance productivity, and develop alternative career strategies. Networking and skill diversification are crucial, as traditional pathways into high-paying jobs may become more competitive. For younger generations, learning coding basics remains valuable, but the ability to work alongside AI is becoming more critical than mastering a single technical skill. Meanwhile, entrepreneurs and those considering career pivots should explore roles that are less susceptible to automation. Investing in assets, such as AI-driven index funds, may provide a financial cushion against potential job losses. While AI’s impact on knowledge-based work is unavoidable, proactively preparing for change can help mitigate the fear of obsolescence.

China’s leading universities are ramping up undergraduate enrollment in key strategic fields, particularly AI, as part of a broader initiative to strengthen national technological capabilities. Institutions like Peking University, Renmin University, and Shanghai Jiao Tong University are adding hundreds of new spots in disciplines such as information science, engineering, biomedicine, and integrated circuits. This expansion aligns with China’s national action plan to become a global education powerhouse by 2035, fostering domestic talent amid growing geopolitical and technological competition. The move is also influenced by the success of AI startup DeepSeek, which has demonstrated China’s ability to develop advanced AI models at a fraction of the cost seen in the U.S. As part of this push, AI education will also be introduced in primary and secondary schools to cultivate digital literacy and innovation skills from an early age.

A breakthrough in AI-powered real estate is reshaping the industry, as a conversational AI agent deployed by Porta da Frente Christie’s in Portugal has facilitated over $100 million in property sales. Developed by tech startup eSelf, this AI system provides real-time, video-driven interactions, ensuring 24/7 responsiveness for potential buyers across different time zones. By personalizing property recommendations and streamlining lead qualification, AI is significantly reducing the time and effort required for transactions. A McKinsey report estimates that generative AI could inject up to $180 billion in value into real estate by optimizing data usage, automating routine processes, and enhancing customer engagement. While AI’s role in real estate is expanding rapidly, challenges such as data governance, compliance, and ethical concerns must be addressed to fully capitalize on its potential. The success of AI-powered property deals signals a transformative shift, positioning early adopters at a competitive advantage in a rapidly evolving digital landscape.

🛠️ AI tools updates

Scientists have developed an AI-powered digital twin model that not only mirrors physical machines but can also control and adapt them in real-time. This breakthrough, called Intelligent Acting Digital Twins (IADT), advances traditional digital twins from passive observers to autonomous decision-makers. Unlike conventional models that analyze and predict system behaviors, IADT learns from human input and autonomously executes actions, making it applicable across industries such as healthcare, smart cities, robotics, and defense. Researchers validated the model using the CupCarbon platform, demonstrating its capability to bridge the gap between virtual and physical systems. By integrating AI and machine learning, IADT represents a step toward autonomous systems that can interact with, adapt to, and manage real-world environments without direct human intervention. This technology could transform industries by enabling machines to operate independently in high-risk and fast-paced scenarios, such as emergency response and autonomous navigation.

💵 Venture Capital updates

ServiceNow has announced its $2.85 billion acquisition of Moveworks, a company specializing in AI-powered enterprise search and self-service solutions. The deal, expected to close in the second half of the year, aims to enhance ServiceNow’s platform with Moveworks’ AI-driven assistant capabilities, providing businesses with more intuitive and automated workflows. This acquisition will allow employees to interact with enterprise systems more efficiently, automating tasks such as IT and HR requests while improving search functionality across organizations. Moveworks, founded in 2016, has developed conversational AI assistants to simplify employee interactions with enterprise systems. Its integration with ServiceNow’s platform is set to accelerate enterprise-wide AI adoption, streamlining workflows and improving productivity for major customers like Siemens, Toyota, and Unilever.

DeepSeek, one of China’s most prominent AI startups, has opted not to take venture capital funding despite significant investor interest. Founder Liang Wenfeng, who holds an 84% stake, has maintained control to prioritize fundamental research over rapid commercialization. DeepSeek has sustained itself financially through profits from Liang’s hedge fund, High-Flyer, reducing its reliance on external capital. Additionally, avoiding Chinese investors helps mitigate concerns about government influence, which has led to bans from certain governments and companies. However, as DeepSeek moves toward monetization and faces chip supply constraints due to U.S. export controls, it may need external funding in the future. While no formal fundraising has been announced, major Chinese tech firms like Tencent and Alibaba have reportedly expressed interest in investing.

🫡 Meme of the day

⭐️ Generative AI image of the day