AI in Schools
After EDUtech: TEQSA's toolkit, eSafety's warning, and the UTS framing
Digital Attitudes, 7 June 2026
The Sunday read is the operational kit. EDUtech wrapped on Thursday, the OAIC code and the Senate inquiry both closed at five on Friday, and the three pieces of paper on the desk over the past five days are the regulator toolkit, the eSafety platform warning, and the academic read of the institutional task.
TEQSA's revised academic integrity toolkit ships AI text detection guides
TEQSA's updated Academic Integrity Toolkit, refreshed on 8 May 2026, adds the operational documents the higher education provider's compliance team needs for the regulatory-led posture the regulator confirmed in May 2025 and codified in the September 2025 Enacting Assessment Reform paper. The May updates include a Good Practice Note, a guide on substantiating contract cheating, a guide on detecting AI in texts, and a guide on the common roles and responsibilities relating to academic integrity. The toolkit also adds two new sections covering assessment design and security and the risks to academic integrity from generative AI. The project team that delivered the toolkit update was led by Associate Professor Guy Curtis at the University of Western Australia. Future Campus on the TEQSA assessment reform pivot the toolkit operationalises.
Why this matters: For vice-chancellors and deputy vice-chancellors (academic) writing the institution's response to the TEQSA regulatory-led framework that started biting this year, the toolkit is the document the provider's compliance team is supposed to be building the assessment design audit against, and the AI text detection guide is the line the institution that still runs Turnitin's AI checker (UNSW Sydney, the University of Queensland, Western Sydney, the University of Tasmania) will read alongside the Curtin January decision to disable the same switch. For Australian higher education CISOs and learning design teams, the guide on substantiating contract cheating is the practical operational document the academic integrity panel will be measured against when the next case lands at the discipline committee. For ed-tech vendors selling AI detection and academic integrity products into Australian universities, the TEQSA toolkit puts the regulator's operational answer to detection on the same page as the regulator's regulatory-led framework, which is the conversation the next renewal call will run into.
eSafety puts WhatsApp and Roblox on notice at Senate Estimates
eSafety Commissioner Julie Inman Grant told a Senate Estimates hearing this week that WhatsApp and Roblox, both currently outside the under-16 social media ban that took effect on 10 December 2025, may be reclassified if they add more social features. The Weekly Times reported the line on 3 June: "the more social you become, the more likely you are to meet the definition." The ten platforms currently inside the ban (YouTube, X, Facebook, Instagram, TikTok, Snapchat, Reddit, Twitch, Threads, Kick) closed 4.7 million under-16 accounts by January, and Inman Grant confirmed at March a 37 per cent reduction in under-16 account holders. WhatsApp is the messaging app the majority of independent and Catholic schools use for parent and carer communication, and Roblox is the gaming platform that has been moving into Australian classroom deployments through the Roblox Education partnership for coding and STEM enrichment. The Weekly Times report on the Senate Estimates testimony, 3 June.
Why this matters: For independent and Catholic school heads running WhatsApp as the parent communication channel, the read is that a regulator-imposed change to the platform's status in the next twelve months would force a procurement decision the school's IT and communications teams have not had to make since the move to WhatsApp from email distribution lists, and the comparator products (Microsoft Teams for Parents, the Compass parent app and the Inklass acquisition Compass folded in earlier this year, the Sentral parent app) are the conversation the next term's communications committee will run. For state Department of Education CIOs and the schools running Roblox Education into Year 5 to 10 coding and STEM units, the operational signal is that the platform's social feature roadmap is now an Australian regulatory question rather than a US product decision. For ed-tech vendors selling messaging, parent communication, and gaming education products into Australian schools, the Inman Grant line is the same line the OAIC Children's Online Privacy Code conversation has been running on the same desks since the consultation opened in March.
Stein in Crikey puts the UTS journalism school read on the record
Ginny Stein, head of discipline for journalism at the University of Technology Sydney's School of Communication, published a piece in Crikey on 2 June arguing that educational institutions need a unified national approach to teach students how to use AI critically rather than be manipulated by it. Stein's central line is that the AI conversation in Australian universities is "causing no end of conniptions across the university sector" and that the institutional task is to help students "confidently, critically and creatively interact with AI." The piece argues for a national approach across institutions rather than the current institution-by-institution policy patchwork that the Australian Framework for AI in Higher Education (ACSES, December 2025) and the Universities Australia response to TEQSA (15 May 2026) have both been trying to thread. Stein's piece in Crikey, 2 June.
Why this matters: For vice-chancellors and deputy vice-chancellors (academic) reading the institution's own AI policy against the ACSES framework and the Universities Australia TEQSA response covered yesterday, the Stein piece is the academic press argument the institution's procurement team will read alongside the TEQSA toolkit update above, because the case for a national approach changes the procurement question from buying institution-specific tools to buying shared infrastructure. For deans of education and heads of journalism, communication, and media schools, the Stein framing puts the AI literacy line on the same page as the curriculum reform line, which is the line the next AITSL submission and the next TEQSA assessment audit will both need to answer. For ed-tech vendors selling AI literacy modules and academic integrity products into the higher education stack, the journalism school read is the sharpest version of the argument that the operational task is teaching, not detecting, which is the same line the TEQSA toolkit update and the EDUtech floor were reading from this week.
The thread today is that the rhetoric is closed and the operational kit has landed. TEQSA's toolkit puts the academic integrity audit on the higher education provider's desk, eSafety's warning puts the platform classification question on the procurement team's desk, and Stein's piece puts the institutional literacy task on the dean's desk. More on the operational read at digitalattitudes.com.au.