**🎉 Today’s Tech & AI News Digest 🎉**
🧠 GLM-5.2 challenges top AI models in cybersecurity
China’s open-weight GLM-5.2 model shows surprising strength in cybersecurity benchmarks, in some cases outperforming Claude and narrowing the gap with leading Western AI systems.
https://semgrep.dev/blog/2026/we-have-mythos-at-home-glm-52-beats-claude-in-our-cyber-benchmarks/
🎓 AI sparks academic integrity crisis at universities
Professors warn that widespread AI-assisted cheating, highlighted by a mass fraud case at Brown University, threatens traditional assessment and trust in higher education.
https://english.elpais.com/education/2026-06-28/ai-fraud-at-brown-university-academic-integrity-is-at-risk.html
⚖️ AI enters the courtroom as digital evidence
Prosecutors’ use of ChatGPT logs in a major arson trial signals how AI interaction data is becoming a new form of legal evidence with profound privacy implications.
https://www.theverge.com/ai-artificial-intelligence/958751/prosecutors-chatgpt-palisades-wildfire-arson-mistrial
🤖 Building language models from scratch
Developers showcase a fully hand-written GPT-2–scale language model in pure C/CUDA, demonstrating deep transparency and control without modern ML frameworks.
https://github.com/JustVugg/nanoeuler
🧪 Can large language models be self-aware?
Researchers debate whether adapted mirror tests meaningfully measure self-recognition or consciousness in LLMs, arguing current approaches are fundamentally flawed.
https://blog.pascalschuster.de/article/do-llms-pass-the-mirror-test
🏭 AI falls short in industrial automation
Ford’s decision to rehire veteran engineers after disappointing AI-driven quality systems highlights the limits of automation in complex manufacturing.
🚗 Tesla’s Full Self-Driving faces growing scrutiny
Regulators and critics intensify their focus on Tesla’s FSD system as AI-driven autonomy continues to raise safety, reliability, and accountability questions.
🩺 Using AI for medical second opinions
A personal account of using Claude to interpret MRI results illustrates both the promise and risks of relying on AI for health-related insights.
https://antoine.fi/mri-analysis-using-claude-code-opus
📜 Proposed KIDS Act could mandate online age checks
New US legislation aimed at protecting children online would require age verification across the internet, raising concerns about privacy and surveillance.
https://www.eff.org/deeplinks/2026/06/kids-act-would-require-age-checks-get-online
💾 AI boom fuels surging demand for memory chips
Wall Street’s enthusiasm for Micron reflects how AI workloads are driving a memory chip supply crunch and reshaping semiconductor market dynamics.
—–
The Daily AI News Digest is brought to you by **#JackHui.com.au**! To know more: [JackHui.com.au](https://jackhui.com.au)