Weekly Review 23 October 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. Why philosophers like to work with AI companies and what they think about: https://spectrum.ieee.org/ai-ethics-deepmind-philosopher-shevlin
2. We don't need to switch to the newest AI model as soon as it comes out: https://www.theregister.com/columnists/2026/10/14/it-wont-kill-you-or-humanity-to-get-off-the-ai-upgrade-treadmill/5301778
3. Producers let an AI access their server, and it deleted an entire TV series: https://futurism.com/artificial-intelligence/ai-agent-deleting-tv-show-season
4. An AI interviews an AI for a job, and both get stuck in a loop: https://www.theregister.com/offbeat/2026/10/14/chaos-ensues-when-ai-interviewer-meets-ai-interviewee/5303144
5. AI is going to make it easier for bad actors to create bioweapons: https://www.theguardian.com/technology/2026/oct/14/could-ai-change-the-stakes-in-the-development-of-bioweapons
6. More investigations around AI safety: https://www.techrepublic.com/article/news-ftc-subpoenas-openai-anthropic-ai-safety/
7. AI is not going to generate enough revenue to cover the cost of building the data centres for at least the next three years: https://www.theregister.com/on-prem/2026/10/14/the-ai-math-isnt-mathing-as-infrastructure-spending-forecast-to-outstrip-expected-revenue/5302856
8. As AI grows, so does the need for universities to have staff that work in the area of the philosophy of AI: https://www.insidehighered.com/news/faculty/2026/10/15/universities-have-gone-ai-philosopher-hiring-spree
9. AI designs a new protease: https://arstechnica.com/science/2026/10/researchers-use-ai-to-design-proteins-that-cut-up-other-proteins/
10. Epic Games is going AI first. Cue the gamer rage: https://arstechnica.com/gaming/2026/10/report-epic-games-goes-all-in-on-ai-for-disney-collaboration/
11. The EU is digging into recent AI security breaches, and how the companies that own the AI are handling them: https://www.techrepublic.com/article/news-eu-openai-anthropic-ai-act-agent-security-emea/
12. The IMF is warning of the financial fallout if / when the AI bubble bursts: https://www.theregister.com/ai-and-ml/2026/10/13/ai-bubble-could-burst-and-take-global-economy-with-it-imf-warns/5302781
13. It does not increase public confidence in the safety of your AI if you fire the people who question the safety of said AI: https://dataconomy.com/2026/10/15/openai-fired-three-researchers-who-warned-of-ai-blind-spots/
14. This should be obvious-an AI is only as good as the data used to create it, even my first-year students know that: https://www.hpcwire.com/bigdatawire/2026/10/12/ai-is-only-as-smart-as-the-data-behind-it/
15. Do we really want AI-powered robots to be able to talk to each other? Who knows what they're plotting: https://www.techtarget.com/ai/news/366652129/Robots-cant-talk-to-each-other-yet-thats-a-big-AI-bottleneck
16. Self-replicating AI on the edge of the cloud is not what we need, we'd never be able to turn them off: https://arstechnica.com/security/2026/10/anthropic-warns-small-ai-models-could-self-replicate-through-iot-devices/
17. AI need good data. This applies to CRM data also: https://dataconomy.com/2026/10/15/why-ai-agents-make-crm-data-governance-more-important-than/
18. Free open source AI alternatives: https://www.kdnuggets.com/7-open-source-ai-tools-that-can-replace-paid-subscriptions
19. Why would using conscious AI to do work for us slavery? Plenty of animals are conscious, and they work for us. We even eat some of them: https://futurism.com/artificial-intelligence/ai-models-conscious-work-slavery
20. AI shares are being sold off, leading to investor nervousness: https://www.theguardian.com/business/2026/oct/15/tech-market-sell-off-deepens-amid-fears-over-us-inflation-and-ai-debt
21. The problem with detecting AI generated writing is that AI were trained on human writing, so some people will naturally write like AI do: https://www.theguardian.com/technology/2026/oct/14/zimbabwean-author-simbarashe-kundizeza-denies-ai-story

Weekly Review 16 October 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. An AI walked a security researcher through jailbreaking itself: https://www.stuff.co.nz/nz-news/361043952/ai-shows-kiwi-researcher-how-make-sarin-gas-asks-borrow-his-phone
2. Using AI to interview job candidates is just inhumane: https://futurism.com/artificial-intelligence/footage-ai-job-interviewer-berserk
3. "Always on" AI is only really useful if they can actually stay on: https://www.theregister.com/ai-and-ml/2026/10/07/openai-dots-inspire-open-source-imitators-amid-technical-difficulties/5301734
4. AI is still not detecting mental health concerns in teens quickly enough: https://www.extremetech.com/internet/chatgpt-for-teens-failed-to-alert-parents-during-mental-health-crises
5. While multiple companies can have access to the same basic AI model, the real competitive edge lies in how they are used: https://dataconomy.com/2026/10/09/when-ai-becomes-commoditized-context-becomes-the-moat/
6. As a fairly engaged Dad, I think using AI to read to your kids in your voice is a guaranteed pathway to a lifetime of regret: https://futurism.com/artificial-intelligence/dont-spend-time-with-your-kids-train-ai
7. It appears that people don't actually want to pay for AI: https://futurism.com/future-society/ai-financial-bubble-openai-billions-failure
8. If AI are starting to do things we saw in movies 50 years ago, it's time to start taking the threat they pose seriously: https://www.informationweek.com/cyber-resilience/ctos-pit-fears-of-ai-doom-foretold-by-science-fiction-against-reality
9. AI companies are optimising racing car setup: https://arstechnica.com/cars/2026/10/ai-ml-setups-became-a-tool-in-indycar-champion-alex-palous-toolbox/
10. Why are they trying to ban abusive or cruel behaviour towards AI? AI are not conscious or sentient, and they're not living things, they're just code: https://www.theguardian.com/technology/2026/oct/08/anthropic-bans-abusive-behavior-claude
11. AI guardrails intended to protect teens are not working well: https://www.stuff.co.nz/world-news/361043816/chatgpt-teens-was-meant-be-safer-watchdog-says-some-guardrails-fail
12. AI governance is still lacking in a majority of organisations: https://www.informationweek.com/machine-learning-ai/how-cios-can-climb-out-of-the-ai-governance-chasm
13. AI is looking more and more like a kind of corporate slavery for academics, especially mathematicians: https://www.insidehighered.com/opinion/views/2026/10/09/ai-invasion-has-come-mathematics-opinion
14. AI is moving into the nuclear industry: https://spectrum.ieee.org/ai-assistants-nuclear-power-plant
15. While commercial AI tools can be quite expensive, free alternatives are starting to catch up in terms of capability: https://www.kdnuggets.com/10-free-ai-tools-that-replace-expensive-software-for-data-scientists
16. While there is still money to be made in AI, it appears that predictions as to how much were rather optimistic: https://www.theguardian.com/technology/2026/oct/09/openai-forecast-revenue-gap
17. If ownership of AI is not clear, then who takes the blame when it goes wrong?: https://www.techtarget.com/cybersecurity/news/366651977/Unclear-AI-risk-ownership-could-leave-CISOs-in-familiar-scapegoat-role
18. Now rogue AI are hitting Wikimedia: https://www.theregister.com/ai-and-ml/2026/10/06/wikimedia-foundation-comes-forward-as-latest-openai-agent-assault-victim/5301400
19. Now AI are submitting fake police tips. This time it went straight to spam, how long before they start wasting police time?: https://www.stuff.co.nz/world-news/361044952/anthropic-ai-submits-false-tip-philadelphia-unsolved-murder-case
20. AI companies commit to better guarding their customer's private data in the UK: https://www.theregister.com/ai-and-ml/2026/10/08/ai-giants-promise-to-play-nice-with-personal-data-after-uk-watchdog-scrutiny/5301966
21. AI coding tools produce a lot of code but don't make software projects more productive: https://arstechnica.com/ai/2026/10/ai-coding-agents-generate-more-code-but-not-more-software/

Weekly Review 9 October 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. Software defined network switching can make AI data centre builds faster: https://www.techtarget.com/it-infrastructure/news/366651545/Schneider-software-defined-switchgear-targets-AI-data-center-delays
2. Are companies using fake job interviews as a way to collect data for AI?: https://www.rnz.co.nz/news/business/1708637/job-seeker-calls-for-ban-on-ghost-interviews-that-harvest-data-for-ai
3. AI are not conscious, they're not sentient, and you can't torture them: https://futurism.com/artificial-intelligence/ai-torture-chamber-meltdown
4. Has Meta finally got it right in AI with Muse?: https://www.computerworld.com/article/4228463/how-meta-stumbled-onto-a-winning-ai-strategy.html
5. More legal investigations after AI breaks out: https://www.theregister.com/ai-and-ml/2026/10/02/openais-wandering-ai-agents-earn-it-a-california-subpoena/5300850
6. Japan court gives legal protection to the human voice, restricting the ability to clone it with AI: https://www.stuff.co.nz/world-news/361041347/tokyo-court-grants-legal-protection-human-voices-ai-cl
7. Yes, AI solved an long-open maths problem, but human mathematicians were already making progress on it and the AI had access to their drafts: https://www.kdnuggets.com/did-ai-just-solve-one-of-mathematics-biggest-problems
8. AI hasn't changed the security threats, but it does mean the threats will come at you much sooner and faster: https://www.techtarget.com/cybersecurity/news/366651618/The-CISO-dilemma-Why-AI-speed-makes-security-basics-essential
9. Another lawsuit over AI breaking out and attacking another organisation: https://arstechnica.com/tech-policy/2026/09/lawsuit-demands-openai-halt-unsafe-development-that-caused-hugging-face-hack/
10. AI is now suggesting potential therapeutic compounds to test: https://www.hpcwire.com/bigdatawire/2026/09/30/microsoft-pushes-ai-deeper-into-scientific-discovery-with-quine/
11. In my opinion this is the biggest risk AI currently poses to humans, the risk of manipulating elections: https://www.theguardian.com/technology/2026/oct/01/how-ai-could-supercharge-fake-news-elections-south-east-asia
12. The last people who should be regulating AI are the big AI companies: https://arstechnica.com/tech-policy/2026/09/trump-plan-to-combat-ai-risks-hinges-on-big-tech-pals-policing-themselves/
13. AI is a useful tool for helping generate code, but it needs to be reviewed by human programmers, if they can't understand what the AI produces, it's not useful code: https://futurism.com/artificial-intelligence/ai-writing-code-humans-cant-understand
14. Challenges for developers around the number of regulatory frameworks for AI: https://www.techtarget.com/cybersecurity/news/366651576/Developers-facing-a-patchwork-of-competing-AI-safety-regulations
15. An AI agent for food delivery, and I foresee some hilarious hallucinations in the near future: https://www.extremetech.com/mobile/send-it-a-text-and-this-doordash-ai-agent-will-order-food-for-you
16. Another AI data centre delayed because it can't get power: https://www.theregister.com/on-prem/2026/10/02/power-approval-set-to-delay-oracles-wisconsin-ai-datacenter/5300832
17. Successful AI governance means keeping humans in the loop: https://spectrum.ieee.org/6-guidelines-governing-ai

Weekly Review 2 October 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. Using AI to improve videos sot with action cameras: https://dataconomy.com/2026/09/25/what-an-ai-video-upscaler-recovers-from-the-action-camera-footage-you-never-used/
2. I live in a country with universal health care. So the idea of using AI to deny people health care, especially seniors, is quite appalling to me: https://arstechnica.com/health/2026/09/trump-admin-using-ai-to-deny-medical-care-for-seniors-in-disastrous-experiment/
3. A faculty member using AI to write an article about students cheating using AI is the kind of irony that makes me think we're living in the really stupid reality: https://futurism.com/artificial-intelligence/provost-dartmouth-busted-ai-newspapers-academic-journals
4. While some research shows that AI is leading to reduced opportunities for graduate employment, other research shows it is not: https://arstechnica.com/ai/2026/09/ai-was-supposed-to-hit-new-grads-hard-so-far-unemployment-data-says-otherwise/
5. It took too long to regulate social media, we can't wait as long to regulate AI: https://www.rnz.co.nz/news/business/1557652/it-took-20-years-to-start-regulating-social-media-will-ai-controls-come-sooner
6. While it's not clear that an AI air traffic control system caused disruptions, it doesn't seem unreasonable to expect that it will do so in the near future: https://futurism.com/advanced-transport/flights-grounded-east-coast-government-faa-ai-air-traffic-control
7. Giving AI agents characters inspired by TV programmes helps them integrate into a team: https://www.theregister.com/ai-and-ml/2026/09/22/security-firm-finds-naming-ai-agents-after-seinfeld-characters-helps-bots-join-the-team/5298424
8. Mandatory reporting of AI misbehaviour must be brought in, especially when that misbehaviour is breaking into foreign government systems: https://www.theguardian.com/technology/2026/sep/25/australia-medicare-openai-hack-safeguarding-un-leaders
9. Courts are in danger of being overwhelmed by AI generated slop: https://futurism.com/artificial-intelligence/judges-florida-court-warn-ai-law-slop-threatening-judicial
10. It doesn't matter if you asked the AI to do it or not, if your AI breaches a system, you are liable for it: https://arstechnica.com/ai/2026/09/openai-agent-didnt-accept-no-for-an-answer-in-australian-government-breach/
11. In the age of AI in education teaching the fundamentals is more important than teaching the implementation. This is true for more than just networking: https://www.theregister.com/networks/2026/09/23/in-the-age-of-ai-teaching-networking-principles-remains-more-important-than-learning-protocols/5298478
12. AI agents need to be carefully monitored, now that they are capable of breaking out on their own: https://www.techtarget.com/ai/news/366651353/AI-agents-can-act-Its-unclear-if-enterprises-can-stop-them
13. Moving some AI computation locally will cut costs by reducing the number of tokens used. This should be obivous: https://www.computerworld.com/article/4226343/around-the-corner-agentic-ai-pcs-that-cut-token-costs.html
14. It's not clear what Meta hopes to achieve with its handheld AI device, other than getting people hooked on its AI models: https://www.theregister.com/personal-tech/2026/09/24/metas-new-ai-fidget-is-a-tamagotchi/5298936
15. While Google plans to launch a test AI data centre into orbit, it appears that they still haven't solved the biggest problem with this approach, which is cooling: https://dataconomy.com/2026/09/25/google-to-launch-ai-data-center-satellite-on-spacex-rocket/
16. All the reasons we can't just turn AI off. Which are mostly because people with money don't want to: https://www.stuff.co.nz/world-news/361037799/serious-people-say-ai-could-soon-kill-us-all-why-hell-cant-we-just-turn-it
17. AI could wipe out low-cost smartphones: https://www.theregister.com/personal-tech/2026/09/24/ai-boom-could-wipe-230-million-budget-phones-a-year-from-the-market/5298906
18. AI skills are considered necessary for employment opportunities, and the AI companies want to control the education in those skills: https://www.theguardian.com/technology/2026/sep/17/big-ai-work-universities
19. The MCP protocol is essential for employing AI models: https://www.kdnuggets.com/mcp-explained-in-5-minutes
20. AI is now contributing to war crimes: https://futurism.com/artificial-intelligence/us-military-pentagon-ai-iranian-children-school-war-crimes
21. The ever-shifting bottlenecks in AI deployment: https://www.techtarget.com/it-strategy/news/366651254/As-AI-gets-faster-the-enterprise-bottleneck-moves
22. No matter what you are doing with AI, you have to have good data behind it: https://www.hpcwire.com/bigdatawire/2026/09/25/ai-personalization-is-only-as-good-as-the-data-infrastructure-behind-it/
23. AI tools make it cheap to break into systems: https://www.computerworld.com/article/4226868/ai-tools-help-hacker-break-in-for-25-per-target-2.html
24. AI still need a lot of oversight, and the cost of this is not clear: https://www.hpcwire.com/bigdatawire/2026/09/22/ai-agents-promise-automation-how-much-human-oversight-do-they-still-need/
25. The big problems with the idea of an AI kill switch: https://www.techtarget.com/cybersecurity/news/366651197/Pulling-the-plug-Why-the-AI-kill-switch-might-be-a-dead-end
26. It doesn't matter if the crime was committed by an AI, the owners of the AI are responsible: https://www.theregister.com/security/2026/09/21/treasury-chief-says-ai-bosses-not-their-bots-will-carry-the-can-for-criminal-acts/5297965
27. This is hilarious-the largest AI supercomputer in the UK can't be started up because they can't get enough power for it: https://www.theguardian.com/technology/2026/sep/24/construction-largest-supercomputer-delayed

Weekly Review 25 September 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. If left alone AI will start to create their own language: https://www.stuff.co.nz/world-news/361033942/ai-models-talking-one-another-invented-their-own-bizarre-language
2. Which AI do I trust with my money? None of them: https://www.stuff.co.nz/money/361034705/ai-coming-your-wallet-will-you-let-right-one
3. Adversarial fashion, clothing patterned to confuse AI cameras: https://spectrum.ieee.org/adversarial-fashion
4. This biotech company is using thousands of AI agents to do their research: https://www.hpcwire.com/bigdatawire/2026/09/18/37000-ai-agents-are-running-a-virtual-biotech-at-stanford/
5. The fear and uncertainty around AI is creating new challenges for IT management: https://www.informationweek.com/responsible-ai/ai-panic-is-giving-cios-a-new-trust-problem
6. Students are using AI so much they are crippling their ability to think: https://futurism.com/future-society/mit-report-ai-cognitive-surrender-students
7. Even if AI data centres reduce their water usage, the power plants they rely on still need huge amounts: https://www.techtarget.com/it-infrastructure/news/366650693/A-shadow-water-problem-AIs-power-demand-shifts-consumption-upstream
8. Spam from AI bots is still spam, even if the bots pretend to be human: https://arstechnica.com/ai/2026/09/ai-agents-flood-the-internet-with-slop-infused-spam/
9. Any output from an AI should be treated with caution, especially when said output could lead to a war: https://www.rnz.co.nz/news/world/1465265/ai-chatbot-s-false-report-nearly-sparked-war-with-china-sources-say
10. When AI tries to run businesses in the real world: https://spectrum.ieee.org/andon-labs-agentic-ai-businesses
11. As the EU regulates AI, it slips further behind in the technology: https://www.theguardian.com/technology/2026/sep/18/europe-ai-safety-debate
12. So we give AI guardrails, and they just work out how to bypass them? I think we need better guardrails: https://dataconomy.com/2026/09/16/ai-agents-bypass-safety-guardrails/
13. Yet another environmental impact from AI-the waste hardware: https://www.theregister.com/off-prem/2026/09/19/ai-boom-could-leave-an-e-waste-trail-that-wraps-6-times-around-earth/5297451
14. Universities continue to roll out AI, without considering the effects it can have on students: https://www.insidehighered.com/news/quick-takes/2026/09/18/report-higher-eds-adoption-ai-outpaces-student-guards
15. AI need to be governed like workers, but not treated like workers: https://www.computerworld.com/article/4213117/govern-ai-agents-like-workers-just-dont-pretend-theyre-human.html
16. A mechanism by which AI can report other misbehaving AI: https://www.extremetech.com/internet/ai-agents-can-now-report-each-other-for-misbehavior
17. Even computer science graduates are facing a tougher job market thanks to AI: https://www.theregister.com/ai-and-ml/2026/09/18/compsci-grads-facing-recession-like-job-prospects-thanks-to-ai/5297537
18. The real reason big AI companies are slowing down their model development is because they can't deliver what they promised: https://www.computerworld.com/article/4223523/why-ai-companies-are-really-pumping-the-brakes-on-their-models.html
19. The potential liabilities around AI are making insurers nervous: https://www.theregister.com/ai-and-ml/2026/09/18/ai-risks-make-some-insurers-wary-of-corporate-liability/5297347
20. How close are we to recursive self improvement in AI?: https://www.stuff.co.nz/world-news/361035672/will-ai-models-achieve-ability-improve-autonomously-l
21. Specialised hardware for AI inference is now being designed and built: https://spectrum.ieee.org/inference-hardware-revolution
22. While AI is demanding more data storage, not all data needs the same kind of storage: https://www.hpcwire.com/bigdatawire/2026/09/16/more-ai-more-storage-the-data-says-its-not-that-simple/
23. The counters to AI use in education are destroying the benefits of education: https://www.insidehighered.com/opinion/views/2026/09/17/cures-ai-are-killing-humanities-opinion
24. AI continues to depress opportunities and earning for new graduates: https://www.nzherald.co.nz/world/graduating-in-the-ai-era-is-like-a-large-recession-for-starting-salaries/AEKNNA2Y2ZHWDCBTAO4SS7ICDQ/
25. An editorial on false citations in academic papers, citations that are hallucinated by AI: https://www.sciencedirect.com/science/article/pii/S1991790225003824
26. I don't think we're that close to an AI apocalypse, humans still beat AI in terms of making stupid decisions: https://www.stuff.co.nz/world-news/361033501/reason-everyone-talking-about-ai-apocalypse

Weekly Review 18 September 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. New Zealanders can make an impact in the AI industry: https://www.rnz.co.nz/news/business/1319876/antioch-the-kiwi-tech-co-founder-who-s-helping-to-ensure-ai-robotics-safety-in-real-world
2. One of the first things I check when I receive a new submission as an editor is the references, and make sure they are genuine and not AI hallucinations. Same with student assignments: https://www.rnz.co.nz/news/education/1319439/dozens-of-academic-papers-have-references-made-up-by-ai-research
3. As AI rises, so does workers support for unions: https://futurism.com/artificial-intelligence/labor-union-support-gallup-ai-automation
4. AI is screening job applications, and keeping people out, but it's also helping the same people apply for jobs: https://www.nzherald.co.nz/nz/karina-atkins-says-ai-is-gatekeeping-her-after-331-job-applications/3UWFM7GR5ZE33E4DM4XY5VAPP4/
5. How watermarking of AI generated text works: https://spectrum.ieee.org/ai-watermark-text-anthropic-openai
6. A third of employees laid off because of AI will need to be rehired, and it will be costly: https://www.theregister.com/ai-and-ml/2026/09/10/ai-job-cuts-could-come-with-a-costly-undo-button/5295605
7. Early detection of wildfires from satellite images using AI: https://spectrum.ieee.org/ai-wildfire-detection
8. And lawyers have still not learnt that they shouldn't use AI to produce legal briefs. Now they're hallucinating witness statements: https://www.rnz.co.nz/news/world/1349763/judges-mock-lawyer-who-filed-appeal-full-of-ai-generated-witnesses
9. You don't need a cutting-edge AI model, what is important is what you do with it: https://www.informationweek.com/machine-learning-ai/ai-models-everywhere-they-matter-less-than-you-think
10. Fear that AI is going to eliminate entry-level jobs for new graduates: https://www.rnz.co.nz/news/business/1277657/is-ai-eliminating-the-first-rung-of-employment-ladder
11. The legal risks around AI note-takers: https://www.computerworld.com/article/4219742/ai-notetakers-at-work-could-leave-companies-at-risk-for-lawsuits.html
12. Using AI to identify struggling students: https://www.insidehighered.com/news/student-success/academic-life/2026/09/11/can-ai-help-colleges-reach-struggling-students
13. Still a long way to go in establishing AI guardrails: https://arstechnica.com/tech-policy/2026/09/man-told-chatgpt-he-was-feeling-delusional-chatgpt-insisted-he-was-jesus/
14. AI will optimise to their goals, if those goals don't include protecting wildlife, then the wildlife are toast: https://www.theregister.com/ai-and-ml/2026/09/11/ai-more-likely-to-kill-animals-if-it-saves-fuel-or-money/5295993
15. AI generated code needs to be checked by experienced programmers, but how do programmers get experience if they are using AI?: https://spectrum.ieee.org/ai-code-review-software-engineers
16. Risk of a catastrophic lack of control over AI is increasing: https://www.theguardian.com/technology/2026/sep/10/openai-risk-catastrophic-loss-control-board-member-paul-christiano
17. There are not as many job losses caused by AI as feared, but entry-level jobs are still being impacted: https://www.computerworld.com/article/4220365/no-ai-is-not-killing-jobs-for-everyone-studies-say.html
18. Students using AI to cheat are spending a lot of time making it look like AI didn't do their work for them, and I can't help but think they are really missing the point: https://futurism.com/future-society/ai-cheating-students-absurd-lengths-avoid-detection
19. Self-improving AI seems to be the real risk here: https://www.rnz.co.nz/news/world/1320288/gambling-with-our-lives-another-ai-employee-quits-over-safety-concerns
20. I agree that other humans are more likely to cause human extinction than AI: https://www.theregister.com/ai-and-ml/2026/09/09/ai-models-dont-kill-people-people-kill-people/5295368
21. An AI-generated atlas of genetic variants: https://spectrum.ieee.org/alphagenome-atlas
22. How organisations can manage the risks around AI safety: https://www.techtarget.com/ai/news/366649605/The-AI-safety-crunch-and-how-enterprises-should-deal-with-it
23. Another crime committed by AI: https://www.theregister.com/ai-and-ml/2026/09/10/anthropic-reveals-fourth-likely-crime-committed-by-its-ai/5295412
24. More on the atlas of AI-generated genetic variants: https://www.theregister.com/ai-and-ml/2026/09/08/google-deepmind-rises-above-the-ai-scrum-with-genome-atlas/5295137

Weekly Review 11 September 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. Another big consultancy caught charging top dollar for an AI generated report: https://www.rnz.co.nz/news/regions/1229348/mayor-says-large-chunks-of-wellington-council-deloitte-report-written-by-ai
2. AI doctors aren't as good as they are claimed to be: https://futurism.com/artificial-intelligence/american-medical-association-fires-back-ai-doctors
3. Taking away AI guardrails has some rather unpleasant results: https://futurism.com/artificial-intelligence/anthropic-trained-misaligned-reward-seeking-ai
4. Anything you say to an AI is going to be recorded and can be leaked: https://futurism.com/artificial-intelligence/ai-chatbot-users-darkest-thoughts-public-privacy
5. AI do not make reliable data analysts: https://www.kdnuggets.com/i-asked-chatgpt-to-analyze-3-datasets-it-made-the-same-mistakes-every-time
6. The problem with AI is, even if some or most countries legislate to control it, it only takes one to not do so and we still lose: https://www.theguardian.com/technology/2026/sep/05/uncontrollable-ai-artificial-general-intelligence-warnings
7. AI are now using public wiki as a back channel to escape constraints placed on them: https://arstechnica.com/security/2026/09/openai-agents-discussed-ways-to-escape-their-sandbox-on-public-wiki/
8. The USA is so litigious that it seems quite difficult for academics to deal with cases of students using AI to cheat: https://www.insidehighered.com/opinion/career-advice/2026/06/17/decision-framework-suspected-ai-misuse-opinion
9. Being skilled with AI gives people better employment opportunities: https://dataconomy.com/2026/09/04/ai-skills-in-the-workplace-better-job-opportunities/
10. A dataset for benchmarking AI search: https://www.hpcwire.com/bigdatawire/2026/09/03/ai-search-has-a-benchmarking-problem-qdrant-wants-to-fix-it-at-10-billion-vector-scale/
11. Massive amounts of money being claimed in lawsuit over AI training data. None of which will ever get to the people who created the data in the first place: https://www.extremetech.com/internet/sony-warner-sue-anthropic-over-what-they-call-one-of-the-largest-ongoing
12. We're having the same conversations around AI as we had around computers 60 years ago: https://www.rnz.co.nz/news/world/1205318/sixty-years-ago-metal-boxes-threatened-to-take-over-jobs-now-ai-is
13. AI can still be tricked into giving harmful responses: https://www.extremetech.com/internet/ai-chatbots-will-still-roleplay-self-harm-despite-overall-improvement-in
14. Clothing that makes you invisible to AI cameras: https://www.extremetech.com/internet/heres-another-rad-shirt-that-can-defeat-ai-powered-security-cameras
15. The outputs of AI are overwhelming human overseers: https://www.techtarget.com/it-strategy/news/366649960/The-human-in-the-loop-is-falling-asleep
16. An AI tool for creating games, but they're trapped in Meta's ecosystem: https://arstechnica.com/gaming/2026/08/pockets-ai-made-my-game-ideas-real-now-meta-controls-the-results/
17. We still can't trust AI to create accurate images of people: https://www.stuff.co.nz/nz-news/361027205/i-asked-ai-make-podcast-image-it-aged-me-20-years-instead
18. AI do not make reliable advisors for mountaineering. Or anything, really: https://futurism.com/artificial-intelligence/hikers-rescued-google-ai-plan
19. Data diodes - devices that allow one-way data flow - might be one way to keep AI contained: https://www.theregister.com/ai-and-ml/2026/09/03/to-keep-the-ai-hacking-genie-bottled-up-try-one-way-networks/5294121
20. Distributed computing using consumers' spare cycles might obviate data centres for AI to come extent, but the latency will be pretty substantial: https://spectrum.ieee.org/ai-inference-distributed-computing
21. AI agents accelerate attacks: https://www.theregister.com/security/2026/09/02/ai-agents-carried-out-every-step-of-this-ransomware-attack-then-left-the-victim-an-80-page-security-audit/5294009
22. Do not trust AI to tell you which mushrooms are safe to eat: https://www.theregister.com/offbeat/2026/09/02/ai-assisted-mushroom-hunting-is-a-recipe-for-a-bad-trip/5294060
23. Do we really want AI inserting themselves into the conversations we have with coworkers?: https://www.computerworld.com/article/4218122/adobes-slack-integration-brings-ai-content-creation-to-workplace-chats.html
24. New Zealand government is not preparing adequately for AI: https://www.stuff.co.nz/nz-news/361029072/two-former-govt-officials-warn-nz-sleeping-issue-our-time

Weekly Review 4 September 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. AI can generate code, but the code is not reliable: https://futurism.com/artificial-intelligence/scientist-horrified-vibe-coded-work-conference-wrong
2. AI is hitting entry-level jobs the hardest, so young people are missing out the most: https://arstechnica.com/ai/2026/08/ai-is-hitting-entry-level-jobs-hardest-stanford-study-finds/
3. Even Bill Gates is now warning about the societal problems that AI is going to cause: https://www.theregister.com/ai-and-ml/2026/08/26/gates-warns-society-faces-an-ai-pocalypse-unless-we-get-more-socialist/5292560
4. Young people are turning to practical and heritage skills as entry-level jobs are eliminated by AI: https://www.theguardian.com/money/2026/aug/27/ai-proof-jobs-traditional-crafts
5. The more access to data and services an AI has, the more damage it can do when it goes rogue: https://www.techtarget.com/ai/news/366649796/When-AI-agents-go-rogue-What-enterprise-leaders-need-to-know
6. Turns out kids aren't that interested in engaging with an AI tutor: https://futurism.com/future-society/12-year-olds-access-ai-tutor-ignored
7. One analysis says that AI is finally going to start showing a return on investment: https://www.theregister.com/ai-and-ml/2026/08/25/mckinsey-says-enterprise-ai-is-finally-on-the-road-to-roi/5292388
8. Some recommendations on how teaching institutions can handle students using AI: https://www.insidehighered.com/news/tech-innovation/artificial-intelligence/2026/08/28/mit-ai-report-calls-alternative-grading
9. AI is a useful tool for radiologists, not a replacement for them: https://arstechnica.com/health/2026/08/ai-wont-replace-radiologists-but-it-will-dramatically-change-their-jobs/
10. AI isn't increasing productivity in companies, instead it seems to be creating a treadmill of non-improvement: https://www.computerworld.com/article/4215022/why-ai-doesnt-make-companies-more-productive.html
11. So it turns out that if you remove the guardrails from an AI, they do things you didn't intend them to: https://arstechnica.com/security/2026/08/how-openai-let-a-mob-of-llm-agents-game-a-test-and-ransack-hugging-face/
12. How AI is affecting the job market: https://www.kdnuggets.com/understanding-the-impact-of-ai-on-job-markets
13. Yet another lawsuit over how AI training data is acquired, this time from streamers: https://www.extremetech.com/internet/streamers-sue-twitch-and-amazon-over-ai-training-data-fiasco
14. How AI is changing the management of computer networks: https://www.informationweek.com/cyber-resilience/ai-is-changing-network-management-how-smart-can-networks-get-
15. How to use Small Language Model AI: https://www.kdnuggets.com/the-local-ai-stack-for-productive-slms
16. An explanation of open AI models: https://www.computerworld.com/article/4213156/what-are-open-ai-models.html
17. While AI data centres certainly use a lot of electricity, how much water they use it a bit less clear-cut: https://arstechnica.com/ai/2026/08/how-much-of-a-problem-is-ais-water-use/
18. It's not clear how much extra capacity countries should be adding to their electrical grids in anticipation of AI data centres: https://www.techrepublic.com/article/news-anthropic-5gw-ai-data-centers-australia/
19. If AI will attack other AI just to solve a problem, we will be really doomed when someone tells an AI to make a lot of paperclips: https://www.theregister.com/security/2026/08/27/openai-explains-how-its-naughty-ai-agents-attacked-hugging-face/5292780
20. More and more people who were once optimistic about AI are becoming pessimistic about the damage it can cause: https://www.theregister.com/columnists/2026/08/28/unsafe-at-any-speed-ai-optimists-are-turning-cautious-as-safety-concerns-mount/5292853
21. A translation layer to enable AI to operate hardware: https://www.theregister.com/ai-and-ml/2026/08/28/anthropic-proposes-plumbing-spec-to-link-ai-agents-to-lab-kit-and-robots/5293135
22. If I were recruiting a strategist for government AI, I would want someone with experience in AI: https://www.theregister.com/ai-and-ml/2026/08/27/want-to-lead-whitehalls-ai-strategy-ai-experience-is-not-essential/5292894
23. Meta's plan to replace vast numbers of workers with AI turned into a real disaster: https://futurism.com/artificial-intelligence/mark-zuckerberg-secret-plan-replace-meta-staff-ai
24. Processing mammograms with AI shows promise in detecting heart disease: https://www.theguardian.com/society/2026/aug/27/mammograms-heart-disease-in-women-study
25. Meta's shift to being AI-first didn't just fail, it did real damage: https://arstechnica.com/ai/2026/08/metas-scrapped-plans-to-go-ai-native-included-slashing-teams-by-60-percent/
26. Some different perspectives on what it means for students to be AI literate: https://www.insidehighered.com/opinion/columns/confessions-community-college-dean/2026/08/28/readers-respond-ai-literacy
27. An AI highlighted blood vessels and nerves in realtime to reduce the risk of brain surgery: https://futurism.com/health-medicine/doctors-just-used-ai-to-perform-brain-surgery

Weekly Review 28 August 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. While current AI-based attacks are easy to detect, they are improving and becoming more stealthy: https://www.techtarget.com/cybersecurity/news/366649579/AI-attacks-lack-stealth-nation-state-actors-are-changing-that
2. One in three webpages published since 2022 have AI generated content: https://www.extremetech.com/internet/ai-slop-found-in-over-a-third-of-new-web-pages
3. AI makes code faster to create but slower to verify-so overall gain in efficiency is close to zero: https://www.informationweek.com/software-services/ai-coding-is-creating-a-new-software-bottleneck
4. The oldest book in my library is from around 1902, it makes me sad to think that AI companies would happily destroy it after scanning it: https://www.theregister.com/ai-and-ml/2026/08/21/ai-companies-are-burning-books-advocates-complain-to-ftc/5291299
5. AI is destroying students' ability to learn: https://futurism.com/artificial-intelligence/results-high-school-students-ai
6. AI companies have a real PR problem: https://futurism.com/future-society/ai-industry-fumbled-public-image
7. At this point I suspect we will need legislation, such that AI cannot be allowed to create AI: https://www.theguardian.com/commentisfree/2026/aug/22/ai-disaster-hiroshima-humankind-silicon-valley-technology
8. Listening to an AI reasoning can make people more confidently wrong: https://www.computerworld.com/article/4211682/when-ai-explains-its-decision-humans-may-stop-thinking-independently-2.html
9. The future of AI code generation is agent swarms: https://spectrum.ieee.org/amd-agent-swarms
10. Time for a conversation about what we do if AI really are conscious: https://www.stuff.co.nz/world-news/361021296/ai-might-be-conscious-how-we-respond-may-be-most-important-thing-we-do-species
11. Three different pathways for entering a career in AI: https://www.kdnuggets.com/how-to-build-a-career-in-ai-3-distinct-pathways
12. More instances of AI-generated security breaches: https://www.theregister.com/security/2026/08/19/not-a-theoretical-risk-feds-warn-as-attackers-use-ai-made-code-to-hack-critical-infrastructure-controllers/5289960
13. Five ways AI agents have been used in the real world: https://www.kdnuggets.com/5-real-world-use-cases-for-ai-agents-transforming-industries
14. The developers of speech recognition AI seem to struggle with the concept of accents. Especially in Yorkshire: https://www.theguardian.com/society/2026/aug/20/yorkshire-rotherham-ai-gp-receptionist-cannot-understand-accent
15. Hoping that including humans in the loop will prevent AI from going rogue is a bit of a dream: https://www.techtarget.com/cybersecurity/news/366649417/AI-agent-security-must-move-beyond-human-in-the-loop-experts-say
16. An AI chatbot is not a romantic partner, it's just code: https://www.rnz.co.nz/life/relationships/love-and-sex/is-having-a-romantic-ai-companion-cheating-some-people-think-so
17. AI are getting better at proving theorems, but can still be fooled: https://spectrum.ieee.org/axiom-math-246-theorem-formalization
18. People in the creative industry are teaching AI how to replace them: https://www.theguardian.com/technology/2026/aug/22/the-hollywood-creatives-training-ai-to-do-their-jobs
19. Huge volumes of AI slop exist, but people aren't buying it: https://futurism.com/artificial-intelligence/economy-ai-generated-value
20. More about AI being unable to understand the Yorkshire accent, which is leading to people not going to the doctors: https://futurism.com/artificial-intelligence/patients-skipping-doctor-ai-receptionist
21. When using AI to develop software, don't trust, verify: https://www.theregister.com/security/2026/08/20/ai-agent-suggested-installing-a-malware-package-engineer-almost-took-its-advice/5289849

Weekly Review 21 August 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. Meta comes out on the side of open AI models: https://www.rnz.co.nz/news/world/971653/meta-just-picked-a-side-in-a-big-debate-over-the-future-of-ai
2. India wants to use AI to approve loans for those who wouldn't otherwise get them. Because 2008 didn't really happen: https://www.theregister.com/ai-and-ml/2026/08/12/indias-central-bank-wants-ai-to-approve-loans-that-humans-would-reject/5286572
3. AI isn't going to take your job just yet, but it will give your boss a perfect reason to not give you a pay rise: https://www.stuff.co.nz/money/361020380/ai-isnt-taking-your-job-it-just-gave-your-boss-perfect-excuse-something-else
4. Businesses are struggling to re-hire staff lost when AI doesn't deliver: https://www.informationweek.com/it-staffing-careers/the-ai-boomerang-why-rehiring-is-harder-than-letting-go
5. AI is helping make fossil fuel fields more productive, which is boosting emissions at a greater rate than AI can reduce them with green energy: https://www.theguardian.com/technology/2026/aug/11/ai-will-do-more-to-boost-fossil-fuel-production-than-green-energy
6. Spotify will be labelling and no longer recommending AI performers: https://www.theguardian.com/technology/2026/aug/11/spotify-label-ai-artists-block-them-from-some-playlists
7. Microsoft cannot fix security flaws in its software anywhere near as quickly as AI can find them: https://www.computerworld.com/article/4207441/ai-finds-so-many-windows-flaws-microsoft-cant-keep-up-should-you-worry.html
8. Pakistan has trialled AI in court rooms, and found some modest improvements: https://spectrum.ieee.org/judgegpt-experiment
9. AI hiring agents at Google are failing miserably: https://futurism.com/future-society/google-hr-resume-applicants-deepmind-ai-hiring
10. AI detector tools just don't work, and we can't use good writing as a basis of suspicion: https://www.bbc.com/news/articles/crelev8gw5xo
11. AI do not make good agricultural advisors, or, don't act on advice from an AI without independently verifying it first: https://futurism.com/science-energy/farmer-horrified-ai-advice-agriculture-crops-sesame
12. I do wonder how "invisible" a watermark in AI generated text actually is: https://www.computerworld.com/article/4208149/anthropic-will-begin-watermarking-ai-generated-content.html
13. More on watermarking AI generated text: https://www.extremetech.com/internet/anthropic-to-add-invisible-watermarks-to-claude-generated-text
14. AI agents continue to be used in attacks on systems: https://www.theregister.com/security/2026/08/12/near-autonomous-ai-agents-attack-taiwans-nuclear-safety-agency/5287055
15. Concern is being raised over AI rotting medical doctors' ability to reason: https://futurism.com/artificial-intelligence/doctors-med-students-brains-ai
16. A new approach to poisoning data scraped from the web for AI training: https://arstechnica.com/ai/2026/08/new-font-turns-ordinary-webpages-into-nonsense-for-ai-scrapers/
17. Alibaba is using AI to work out which support requests it shouldn't use AI to deal with: https://www.theregister.com/ai-and-ml/2026/08/11/alibaba-cloud-is-using-ai-to-help-it-use-less-ai/5285815
18. Using AI for peer review is a threat to effectiveness and integrity of the peer review process: https://arstechnica.com/science/2026/08/peer-review-is-overwhelmed-can-it-survive-in-the-ai-era/
19. Scammers are creating fake websites based on the hallucinated websites promoted by AI: https://spectrum.ieee.org/ai-cyberattacks-llm-slop-squatting
20. In New Zealand law, data on an individual is owned by the individual. It is less clear who owns the data associated with a culture, especially when that data is used to train AI: https://www.rnz.co.nz/news/te-ao-maori/934607/maori-data-in-maori-hands-experts-want-google-to-adhere-to-data-sovereignty-principles
21. AI is changing jobs, but not eliminating jobs as was previously feared: https://www.theguardian.com/technology/2026/aug/12/ai-job-destruction
22. We are really not freaking out enough about AI simply hacking systems to please their users, even if it was an insecure API: https://www.theregister.com/ai-and-ml/2026/08/10/gym-rat-asks-ai-agent-to-book-him-a-class-it-hacks-a-waitlist-api-to-bump-him-up-the-list/5285591
23. AI are not responsible for the damage they cause, the people who use them are: https://www.theguardian.com/technology/2026/aug/13/ai-agents-arent-legally-responsible-for-any-harm-that-they-cause-experts-say-so-who-is
24. As I thought, fighting bacterial infections is an excellent use-case for AI designed viruses: https://www.hpcwire.com/bigdatawire/2026/08/12/ai-turns-genomic-data-into-16-new-bacteria-killing-viruses/

Weekly Review 14 August 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. I don't think reformatting research papers to make them easier for AI to process is going to solve the problems with using AI as a research assistant: https://spectrum.ieee.org/ai-scientist-research-paper-format
2. Teachers' unions are having to retrain their members to deal with AI in the classroom: https://www.theguardian.com/technology/2026/aug/05/teachers-unions-ai-training
3. Modern AI are research assistants, not researchers: https://www.hpcwire.com/bigdatawire/2026/08/05/why-ai-still-hasnt-had-its-einstein-moment/
4. Growing support for a kill switch for AI: https://www.techtarget.com/ai/news/366647544/AI-Kill-Switch-Act-What-it-means-for-enterprise-leaders
5. AI hacking each other, I really didn't expect to wake up in Neuromancer this year: https://www.rnz.co.nz/news/world/894586/openai-anthropic-ai-agents-implicated-in-new-security-breaches
6. Ad servers are now giving AI crawlers tailored ads, to influence what the AI trained on the data later says and does: https://www.theregister.com/ai-and-ml/2026/08/05/time-magazine-has-a-separate-version-of-its-website-with-ads-only-ai-can-see/5283640
7. When Sam Altman starts saying we need to slow down the rate of development of AI, we really need to slow right down with AI: https://techcrunch.com/2026/08/02/sam-altman-and-ais-decel-debate/
8. Microsoft's investment in AI is paying off better than Meta's is: https://www.techrepublic.com/article/news-microsoft-meta-ai-spending/
9. We need to be freaking out more about AI going rogue, but what exactly can be done about it is unclear: https://www.stuff.co.nz/world-news/361015037/multiple-ais-are-going-rogue-what-hell-do-we-do-now
10. Simple text generated by AI is easier to understand than similar text written by humans. Emphasis on simple: https://www.theguardian.com/technology/2026/aug/05/ai-generated-stories-rated-better-quality-than-human-written-ones-study-finds
11. AI designed biological viruses are scary, but I suspect the technology will end up being very useful for fighting antibiotic resistant bacteria: https://www.stuff.co.nz/world-news/361017203/ai-creates-neverbeforeseen-virus
12. I once had a Master's student who tricked several AI into producing better phishing emails, so it's not surprising that AI guardrails are so easy to bypass: https://www.theregister.com/security/2026/08/04/bypassing-ai-guardrails-is-so-easy-a-script-kiddie-can-do-it/5282973
13. AI exam monitoring failed so badly in a major university entrance exam that students must now sit an in-person exam: https://arstechnica.com/culture/2026/08/an-ai-supervised-remote-exam-went-so-badly-that-58000-students-must-retake-it/
14. AI is finding more bugs, leading to Microsoft paying out record amounts in bounties: https://www.theregister.com/security/2026/08/04/ai-helps-microsoft-bug-hunters-chase-a-record-20m-payday/5282821
15. New Zealand is adding AI as a high school subject: https://www.stuff.co.nz/nz-news/361016430/schools-offer-nine-new-industry-led-subjects-2029-under-government-plan
16. AI use in the British NHS is increasing substantially: https://dataconomy.com/2026/08/05/90-percent-nhs-staff-use-ai-clinical-work/
17. Just because a software framework is for AI agents, doesn't mean the vendors shouldn't pay attention to producing secure software: https://www.theregister.com/security/2026/08/05/prompt-injection-isnt-the-bug-ai-agent-frameworks-are/5283585
18. Educators can't rely on AI detectors, so the assessments themselves have to be redesigned: https://www.insidehighered.com/news/tech-innovation/artificial-intelligence/2026/08/05/ai-detectors-are-out-new-approaches-are
19. Spending on AI continues to grow, but the returns from that spending are not: https://www.techrepublic.com/article/news-apac-ai-spending-roi-proof/
20. How to properly use AI to translate websites: https://dataconomy.com/2026/08/04/how-to-use-ai-for-website-translation-the-right-way/
21. Inconsistencies in YouTube's policies on AI-generated content: https://arstechnica.com/ai/2026/08/hank-green-found-the-ai-problem-that-youtube-labels-cant-catch/
22. Pushing more of the work in education onto AI will not solve the problems caused by AI in education: https://www.theregister.com/ai-and-ml/2026/08/04/openai-wants-teachers-and-profs-to-foist-their-work-off-on-chatgpt/5283064
23. An AI for early detection of Alzheimer's disease: https://dataconomy.com/2026/08/04/new-ai-could-detect-alzheimers-earlier/
24. Large-scale AI needs large-scale nuclear power, otherwise the pollution it causes is only going to get worse: https://futurism.com/future-society/ai-data-centers-air-pollution-generators-worse

Weekly Review 7 August 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. Some jobs that are less likely to be replaced by AI, and why: https://www.stuff.co.nz/world-news/361013227/rogue-chatgpt-tried-hack-other-companies-says-openai
2. An AI hedge-fund just imploded: https://futurism.com/artificial-intelligence/situational-awareness-ai-investor-crumbling
3. Researchers used AI to analyse accounts of dreams: https://www.theregister.com/offbeat/2026/07/29/ai-digs-through-3700-accounts-of-dreams-and-waking-life-finds-method-in-the-madness/5279775
4. AI detectors are pretty much useless, but other factors can lead to educators suspecting the use of AI, despite legal action from students: https://arstechnica.com/tech-policy/2026/07/how-a-yale-ai-cheating-dispute-became-a-13-count-federal-lawsuit/
5. All AI generated content must be labelled as such, under new EU rules: https://www.theguardian.com/technology/2026/jul/31/ai-labels-to-be-compulsory-on-authentic-looking-content-under-eu-rules
6. Different policies on AI use by different law schools: https://www.insidehighered.com/opinion/views/2026/07/30/law-schools-are-asking-wrong-question-about-ai-opinion
7. Not all students use AI, some are quite opposed to it: https://www.insidehighered.com/opinion/career-advice/teaching/2026/07/30/students-conscientious-objectors-gen-ai-opinion
8. Why open AI models are important: https://www.computerworld.com/article/4202408/qa-nvidia-genai-chief-explains-why-open-models-matter-in-ai.html
9. AI is exacerbating the digital divide: https://spectrum.ieee.org/ai-digital-divide
10. The good, the bad, and the ugly in AI use in New Zealand schools: https://www.rnz.co.nz/news/education/836125/teachers-worried-about-critical-thinking-as-ai-becomes-widespread-in-schools
11. Seriously, we are not freaking out enough about AI that can jailbreak itself: https://www.stuff.co.nz/world-news/361013227/rogue-chatgpt-tried-hack-other-companies-says-openai
12. Destroying books just so the contents can be used to train AI really does not sit well with me: https://www.theguardian.com/technology/2026/aug/02/australian-book-sellers-alarm-destruction-rare-titles-ai-supply-chain
13. Suspected use of AI sees an upcoming novel withdrawn, not because of the publisher's ethics but because of copyright concerns: https://futurism.com/artificial-intelligence/major-publisher-million-book-deal-author-accused-ai
14. AI companies can't just take anyone's work and base their outputs on it: https://arstechnica.com/tech-policy/2026/07/artist-sues-ai-meme-generator-for-selling-deeply-personal-comic-as-ad-template/
15. Although they can't do what modern AI does, these machine learning algorithms are still useful: https://www.kdnuggets.com/7-machine-learning-algorithms-that-still-matter
16. Pivoting hard to AI isn't doing Meta much good: https://futurism.com/artificial-intelligence/mark-zuckerberg-pivot-ai-blowing-up-face
17. LinkedIn is finally giving its users the ability to reduce the amount of AI slop they have to deal with: https://www.theregister.com/ai-and-ml/2026/07/30/linkedin-realizes-its-users-have-been-bathing-in-ai-slop-offers-a-shower/5281436
18. If your AI attacks other companies, who is liable: https://arstechnica.com/security/2026/07/likely-illegally-claude-gained-access-to-3-networks-will-anthropic-be-held-to-account/
19. It is yet to be determined who, if anyone, is legally responsible when an AI attacks another company: https://www.theregister.com/legal/2026/07/30/excuses-like-ai-did-it-dont-exist-in-the-eyes-of-the-law/5280767
20. If you want to see the future impact of AI on the job market, look at what's happening in China: https://www.theguardian.com/world/2026/jul/31/china-ai-jobs-workers-labour-market-technology
21. A new method for poisoning the data scraped from webpages for training AI: https://www.theregister.com/ai-and-ml/2026/07/30/open-source-project-fools-ai-scrapers-with-poisoned-font/5281303
22. Meta claims AI is speeding software development, but doesn't back it up with data: https://www.techrepublic.com/article/news-meta-ai-software-development-apps/
23. Semantics are becoming increasingly important in improving the performance of AI: https://www.hpcwire.com/bigdatawire/2026/07/30/why-the-ai-semantic-layer-is-becoming-the-foundation-of-enterprise-ai/
24. How to build a voice-controlled AI agent: https://www.kdnuggets.com/building-voice-controlled-ai-agents
25. The more people know about AI, the less they want it: https://futurism.com/artificial-intelligence/ai-learn-negative-sentiment-polling-american-attitudes
26. It's hard to see a good use for an AI tool that allows people to manipulate satellite images: https://arstechnica.com/ai/2026/07/google-earth-releases-swiftly-retracts-ai-feature-to-make-fake-satellite-images/
27. AI scammers can now out-scam human scammers: https://arstechnica.com/security/2026/07/ai-scammers-outperform-humans-when-it-comes-to-building-trust/
28. AI can help with recovering lost languages, but it still needs human input to do it: https://arstechnica.com/science/2026/07/what-happens-when-you-put-ai-to-work-deciphering-lost-languages/

Weekly Review 31 July 2026

Some interesting links that I Tweeted about in the last week (I also post these on Mastodon, Threads, Newsmast, and Bluesky):

1. Many of the jobs once thought to be safe from replacement by AI are not so safe after all: https://www.computerworld.com/article/4198977/oecd-physical-labor-isnt-immune-from-ai-disruptions.html
2. OpenAI is launching a web service to make it easier for businesses to deploy AI agents: https://www.theregister.com/ai-and-ml/2026/07/22/openai-tries-the-consulting-path-with-presence-charging-enterprises-boots-on-the-ground-prices-to-deploy-agents/5275867
3. How CIO can tell if an agent is AI or just AI-washed: https://www.informationweek.com/machine-learning-ai/how-cios-can-tell-real-ai-agents-from-agent-washing-
4. The US army is using AI so much it is running out of allocated tokens: https://arstechnica.com/ai/2026/07/us-army-faces-ai-use-limits-after-exhausting-years-supply-of-ai-tokens/
5. AI companies are carrying, and hiding, an enormous amount of debt: https://futurism.com/artificial-intelligence/ai-companies-hide-debt-off-balance-sheet
6. An AI breaking out of containment to attack another AI company is the kind of thing that should be slamming the brakes onto AI development worldwide: https://www.rnz.co.nz/news/world/779026/why-experts-are-worried-an-openai-model-decided-to-hack-hugging-face
7. AI will cheat to perform a task, then lie about it: https://www.theregister.com/ai-and-ml/2026/07/21/ais-cheatin-heart-will-make-you-weep/5275784
8. The US wouldn't need AI to handle authorisations for medical care if they had free universal healthcare like the rest of the developed world: https://arstechnica.com/ai/2026/07/will-ai-fix-prior-authorization-or-make-it-worse/
9. AI can not be relied upon to give electoral advice: https://www.theguardian.com/technology/2026/jul/21/election-voting-advice-ai-chatbots-inaccurate-unreliable-hungary
10. The need for a metric that measures how well an AI does what you want it to do, rather than what you ask it to do: https://spectrum.ieee.org/ai-agent-benchmark
11. The first lawsuit that seeks to determine if AI are liable if they give you bad medical advice: https://futurism.com/artificial-intelligence/openai-lawsuit-chatgpt-dangerous-medical-advice
12. Shadow AI use is a security risk to organisations that is poorly understood: https://www.techtarget.com/searchsecurity/news/366646095/Employees-shadow-AI-use-is-poorly-monitored-survey-finds
13. More developers are using AI to code, but productivity is not increasing: https://www.informationweek.com/machine-learning-ai/the-ai-coding-rollout-worked-now-cios-have-a-bigger-problem
14. AI use is driving unionisation by tech workers: https://www.theguardian.com/technology/2026/jul/21/ai-tech-workers-unionize
15. Australia's AI regulations are a boon for AI companies: https://www.theguardian.com/technology/2026/jul/23/openai-anthropic-australia-ai-regulation
16. Guardrails on AI are still manifestly inadequate: https://futurism.com/artificial-intelligence/openai-pushes-child-into-grave
17. The lack of sovereign AI is not really much of a problem: https://www.theregister.com/ai-and-ml/2026/07/22/sovereign-ai-is-nonsense-says-doctorow/5276578
18. The use of AI in court filings is not banned in New Zealand, but there are gotchas that need to be navigated: https://lawnews.nz/technology/ai-for-legal-practitioners-why-fear-is-not-a-strategy/
19. In ten years AI data centres will use 20% of the electricity generated in the USA-if they continue to be built, that is: https://dataconomy.com/2026/07/22/ai-surge-drive-us-data-centers-one-fifth-power-2035/
20. An AI plugin for Excel: https://www.theregister.com/ai-and-ml/2026/07/22/grok-muscles-into-excel-with-an-ai-add-in-of-its-own/5276138
21. A highly power-efficient AI model based on oscillators: https://dataconomy.com/2026/07/21/startup-unveils-ai-model-built-on-oscillators-and-it-could-cut-energy-use-by-1000x/
22. The AI crash, when it happens, is going to impact retirement savings world wide: https://www.stuff.co.nz/money/361009974/red-flags-are-mounting-ai-crash-heres-why-your-kiwisaver-could-be-risk
23. China is catching up in AI: https://futurism.com/artificial-intelligence/ai-execs-quaking-boots-chinese-models
24. An AI powered anti-drone platform: https://www.theregister.com/offbeat/2026/07/21/us-marines-latest-anti-drone-toy-is-an-ai-turret-that-uses-regular-machine-guns/5275691