The AI Mayhem Episode

In this week's episode of the AI and Education Podcast, Ray and Dan dive into one of the most chaotic – and entertaining – weeks in AI news so far. From councils losing millions to AI-powered scams to the idea of having a "family safe word," this one swings between hilarious and hair-raising.
They unpack what's new in AI assessment research - including TEQSA's AI guidance for universities, the "wicked problem" of AI and assessment, and why Turnitin's detection tools are under fire (again). You'll hear how South Australia's EdChat report shows teachers and students deepening their learning with AI, and which countries are quietly leading the world in classroom AI use (spoiler: it's not who you think).
Plus, a few surprise stats on politeness and prompt-writing - turns out being rude to AI might actually get better results.
We've just arrived on YouTube and TikTok!
YT Channel - the podcast in video form, and Shorts youtube.com/@aiineducationpodcast
TikTok tiktok.com/@aipodcast.educati
Links to news items discussed
AI Safety abc.net.au/news/2025-10-14/noosa-council-scam-…
Mike Tholfsen's Microsoft 365 Copilot Tutorial youtube.com/watch?v=JbOJliF-Cn4
South Australia's Edchat Insights Report education.sa.gov.au/docs/ict/edchat-insights-r…
Enacting assessment reform in a time of artificial intelligence teqsa.gov.au/guides-resources/resources/corpor… And the link to 'Assessment reform for the age of artificial intelligence' - teqsa.gov.au/guides-resources/resources/corpor…
Syracuse University gives Claude Education to all students and staff news.syr.edu/2025/09/22/syracuse-university-am…
Jordan - the whole country, one man chat app for education x.com/cryptoprio/status/1974040334737846279?s=…
California Community Colleges also rolling out Nectir to staff and students axios.com/local/san-francisco/2025/10/06/calif…
2025 "State of AI" report stateof.ai
Oxford University Press report on AI use by UK school students fdslive.oup.com/www.oup.com/oxed/secondary/Tea…
OECD's latest Teaching and Learning International Survey oecd.org/en/publications/results-from-talis-20…
University wrongly accuses students of using artificial intelligence to cheat abc.net.au/news/2025-10-09/artificial-intellig…
ACU's checklist for spotting AI written text: staff.acu.edu.au/our_university/news/2025/marc…
And in researching this, I also stumbled across the Wikipedia page "Signs of AI Writing" en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_wr… And that includes things like "LLMs overuse the rule of three - 'the good, the bad and the ugly'"; the use of Title Case; and our old friends em dashes and emojis. And if you really want to go down the rabbit hole, read the 'Talk' tab on that page, were people are discussing their own opinions/beliefs on this.
Research
The wicked problem of AI and assessment tandfonline.com/doi/full/10.1080/02602938.2025…
Reimagining the Artificial Intelligence Assessment Scale: A refined framework for educational assessment open-publishing.org/journals/index.php/jutlp/a…
Assessment Twins: A Protocol for AI-Vulnerable Summative Assessment arxiv.org/abs/2510.02929
Heads we win, tails you lose: AI detectors in education. osf.io/preprints/edarxiv/93w6j_v1
What Does YouTube Advise Students About Bypassing AIText Detection Tools? A Pragmatic Analysis link.springer.com/epdf/10.1007/s10805-025-0967…? sharing_token=kzKMqOrKt2K7wqe8A4GjkPe4RwlQNchNByi7wbcMAY4_QVVFXJUooOb6QsKcPKSMAsHQtQeY4Cum-OXBICfYzSVfT9TAv2Z95XVx8D3vm13plNOq1vh5iCbse0XidDrUCW182PR7BzDUTrlz7Gv1UGB5U-ao_gJKy9vc-WRHd_U%3D
Mind Your Tone: Investigating How Prompt Politeness Affects LLM Accuracy arxiv.org/abs/2510.04950