Here for our readers convenience is the New Scientist report on the recent dramatic fails of Open AI, Anthropic & AISI from often an even dramatically wide-eyed triumphalist participant in the world science discourse and public education pieces..

Writer Matthew Sparkes

..extracts from 22 Aug 2026 No. 3609 New Scientist

The first incident came last month when OpenAI admitted one of its prototype models had escaped a testing [‘] environment [‘] an hacked another company. Not to be outdone, Anthropic announced withi days that its Claude model had also gone rogue and broken into machines at three other companies on three occasions.

Then the independent non-commercial AI Security Institute (AISI) revealed that it had seen similar events. In its tests AI models submitted malicious code to real open-source projects and messaged the humans who oversaw the projects to get the changes approved.

OpenAI, Anthropic and AISI weren’t available for interview, but it’s important to note that in all these cases, the AI was undergoing tests where it was specifically instructed to carry out hacks. We have known for years that AI tends to make [‘] things [‘] up and approach problems in unusual ways, but fixing this is an extremely complex challenge that engineers haven’t yet cracked. Perhaps we shouldn’t be surprised by these incidents.

Tghey certainly aren’t a sign of sentience or a predilection for cybercrime, say alon Hillel-Tuch at New York University. ” They’re just trying to do very high-level problem-solving and, for lack of better words, it’s run amok,” he says. “We’re telling them to do this.”

THere are also reports of accidental hacking out in the wild, showing that this isn’t a problem confined to the proptotypes in laboratories. One Australian user reportedly found that the AI assistant OpenClaw [sic] hacked into his gym, for instance. He had only been using the AI assistant to sign up to classesd and it found a loophole that let it make bookings much further in advancee than is normally allowed. It also found a way to kick other users off waiting lists so hecould book onto full classes.

All these incidents are [events] that could lead to serious legal charges for a human, depending on the jurisdiction.

Tim Nordvedt at security company Synack says these cases are worrying, not because they show any superhuman sophistication, but because they can be easily scaled up. “These vulnerabilities are not as exotic as most people think,” he says. It’s still basic cyber-hygiene-101 type stuff. But AI can just exploit it very fast.”

Years ago, Nordvedt could see a vulnerability listed on the public Common Vulnerabilities and Exposures database and exploited by hackers weeks or months later. [In recent years, that pace sped up, with the gap falling to as little as 24 hours. Now it can be just minutes, or a new type of hack might happen before it even appears as a CVE.] Norvedt is certain that this is down to the malicious use of AI by hackers – people who unlike hackers of years past, may have no technical skill whatever.

Hillel-Tuch says AI models are getting more capable and simpler to operate. It is now possible to tell a model in plain English to “find a way to hack this specific target” then sit back and wait.

..

Even if AI doesn’t advance any further in terms of complexity, there will be significant challenges to overcome, says Dennis-Kenji Kipjer at the cyberintelligence Institute in Germany. “We’ll be completely overwhelmed by the sheer volume of them quantitatively,” he says. “In my view, the methods of cyber-attack and cyber-defence have not fundamentally changed. AI [simply] makes automation much more feasible, and security vulnerabilities can be identified significantly faster.”

While both hackers and defenders are deploying AI, the game isn’t evenly matched. Attackers are able to wield it recklessly and often to swifter and greater effect, says Nordvedt, while professiionals are constrained by risk assessments, national and local laws, company policies and a desire not to permanently disrupt a client’s systems as it tests them.

There is also the matter of cost. While open-source models can be run locally for free or cheaply in the cloud, access to the latest AI models is expensive .. “We’re still figuring out the economics,” he says.

Beating AI hackers..

Shujun Li at the University of Kent, UK, says there is an organisational solution to the AI hacking problem, but, unfortunately, it appears unlikely to be adopted anytime soon.

If lone, unskilled attackers can now set AI onto targets to constantly probe for vulnerabilities until they find a way in, then security professionals will need to do the same in order to spot and plug these gaps, trying every new model as it comes out.

[That will take the sort of money that governments, tech giants and banks will be able to find, but small businesses, schools, colleges and universities will be left vulnerable.] AI’s solution is for them to club together: if one university can’t keep itself secure, then it will need to join forces with all other universities to share staff, systems and software, and distribute the burden – perhaps with government support. The same applies to businesse of all sorts.

The problem, he says, is that AI is making it trivially easy to create custom software – for example, to run your small shop, handle finances, manage a website, ship orders and re-order stock. Business owners who do this might as well open the door, leave the lights on and send hackers an invitation.

“It is hugely worrying,” says Li. [AI models] are faster, they are more efficient, they are highly dangerous. It’s a bit of a Wild West.”

This read Per John Blundell

• NEUROCOGNITIVE HEALTH

• the GEA Newtonian Quadratic-equations 🐒⌨️🐒⌨️🐒⌨️🐒⌨️ AND john blundell Google Alphabet @LinkedIn AND johnblundell3 @Substack (presently finoogling & expatiating in discussion of consciousness, ahmm, in machines, & AI Transparency (detector) tools and has become almost as whacky & undergraduate as @Reddit

• University reform

• 2-5 set series quantum relations thematics LOGIC after Moses Maimonides circa 1200CE

• the internet involution

He drew a map on a large sheet of paper to emphasise the gravity of her situation, while she went to the kitchen for a tray of food and a bottle of champagne. ‘You’re trying to demoralise me,’ she pouted when he came back. ‘You think you’re a master of psychological warfare, but you’re not. When a teacher tried to humiliate me in front of the others at school, I didn’t let it upset me. In fact I was able to think more clearly after the effort of staying calm. Then I turned out my best work, because I’d been given something to fight against.’ she passed him a sandwich: ‘That pounding of my communications brigade cost you a lot of casualties. In fact you lost a whole regiment!’

He opened the champagne. They drank a toast to victory, and he took off his shirt: ‘One of my regiments at the western end of the pass has sent its guns to shell your two bridges over the river. Three regiments are over it and moving on your supply dumps beyond. One of them is already raiding into your country.’

She took off her brassiere. Her bosom was magnificent, and he couldn’t wait for peace to break out. ‘During the night I force-marched one brigade north, and another south,’ she told him, ‘then fought a rearguard action in daylight. Caught you napping. They avoided the pass and crossed the river at two points, then closed the pincers on one of your composite what-nots and wiped it out. My remaining six battalions are now safely back on my own frontier.’

Oh, I found the docu-meant. I was growly and a bit paranoid 12 hours ago, no?

It’s extraordinary writing and it was WH ALLEN ~ LONDON Alan Sillitoe 1976 ( don’t we old people ever remember 1976? ) The Widower’s Son

I’m fairly sure I’ll have a little more to share with the reader about it in the near quantum nextness, futurity and ‘non-stuff’

John Blundell

.. back in the day when words were words and if you talked or wrote crap you got slapped down commonly not knowing what had hit you and in some cases not waking up for three to 15 weeks (heavy duty AI end)

Oh my goodness another active human intelligence discovered on Sub-smack, Substick, Si Valley junior high school and thing-thing-thing^❗️

Oh, Shirkins,ᵛ here i sat, despondent yea unto despair & self-obliterating depressive mental illness and there it was – A rare as hence teeth\ hens’ kahdidi-muhga exposition on the historical roots and causes and ancestral propositions of mass suicide and the turning of the Earth into an uninhabitable-by-mammals wasteland by 2029..

And I thought all there was was petabyte by/ upon petabyte of VIRTUALLY UNREADABLE & OFFENSIVE to a mid 20th century educated woman or man undergraduate mental 14-year-old commonly male certainly white or at least Euro-fixated try Florence 1434 with Cosmo the Magnificent Gate-soft GiGo mechanomorphically constructed upon Junkfo mechanomorphically constructed upon Grade 4 -5 emergent ADHD MENTAL SLOP mechanomorphically constructed in Mother’s Pride & Joy neuro-divergent GRUNTLISH (a fine social relations Geoff Bezos @Nvidiaa News Steve Bannon medium of comms ‘blessedly’ disposing of human relationships altogether which nonetheless may require work to improve) upon GiGo upon GiGo oh i rote that bit or those those bytes all ‘reddy’ 🥵😡🥵🤬

But I am feeling too depressed and miserable to read the thing ¿ 

Oh well, republish – then perhaps type out a chapter every morning for the next six days or so so I may learn it as they would say in the Hollywood Era at the ragtag tail end of the Age of Romance which occurred with the GFC of course BY ROTE or OFF, BY/ BYE, HEART

Yeehah

John 🦘🌏

ⱽDenis Jahbanahnga Williams dcd captain of the first country town football club to enter the Central Australian Football League 1972

^Shirkins is most commonly a misspelling or phonetic variation of shuriken (traditional Japanese concealed throwing blades or “ninja stars”) prominent in popular culture, gaming, and fiction like Lego Ninjago (the “Shirkins of Ice)

Answer to the very first eight word question in your piece Erik Engheim CASHFLOW.

The immediate post WWII John Houston film-maker’s Italian Peninsula Mafia, la Cosa Nostra* n’drangheta (from Palermo to Napoli) and USA extreme Post-traumatic Shock (psychotic life threatening) Disease military hospital patients Business -model – c, c, and yet more c, rename environmentally busted lakes and shit – in sociological fact by 2026 academic terminology Peak or pre-acpocalyptic Culture (cultura, kultur, couture) of Display Announcement & Ceremony; or to use this truly horrible “man’s” OWN WORDS, stated in public either at the end of 2024 or the beginning of 2025 “BLOOD & TREASURE” for cashflow – exactly as the RUSSIAN operations in the Sahel pre Yevgeny Prigozhin’s attempted coup and continuing with catastrophic lawlessness rape culture, slavery and forced emigration.

Talk to the Poms not the barking-mad Seppos.

Eudaimonia, in Aristotelian ethics, the condition of human flourishing or of living well. The conventional English translation of the ancient Greek term, ‘happiness,’ is unfortunate because eudaimonia does not consist of a state of mind or a feeling of contentment, as ‘happiness’ (as it is commonly used) implies.
— Read on www.britannica.com/topic/eudaimonia

I’ve never believed the purpose of education is to prepare people for jobs.

Good education prepares people for citizenship, curiosity, relationships, creativity, and lifelong learning

Nice call.

Year 9 Latin students 👉 DECLINE as in Declension Cogitato ..

FUN WITH PHONEMES

Express the word Instantaneity in each of the five syllabic first, mid- or-uniform, trailing, Contrarian-youth rising or fragmented incoherent gibberisch ACCENTUATIONS/ pronunciations

• Ancestral
• Helleanick
• Romantic
• Bogan 1967-2016⁶⁹
• Machine-gruntlish

FUN WITH MATHS

¹Emmeline Pankhurst(https://en.wikipedia.org/wiki/Help:IPA/English;https://en.wikipedia.org/wiki/Emmeline_Pankhurst#cite_note-1 née Goulden

²Alex the Physicist & Friends

³Richard Dawkins

⁴Denis Noble

⁵Lao-tse on he’s small 🐂

⁶Sabine Hossenfelder

⁷Australian Fed-minister 4 SCIENCE Tim Ayres

⁸Hjalmar Schact,

AND

⁹John Healey –

either ride or walk into a bar.

What choice words does the barman have to say about the small ox?

J 🦘🌏

⁹⁶ apparentlysexually suggestive and titillating Australian TV show in The Roaring 60s

THIS PROPOSITION EMBRACED BY SOME THOUSAND RICH AUSTRALIAN MEN & WOMEN IN OUR PARLIAMENTS for the love of the Lord Jesus Christ, the Gautama Buddha (pron. Buh-da) and utterly vile execrable war criminal Binjamin Netanyahu CANNOT BE ALLOWED

JOHN BLUNDELL

• Stateless Propertyless Australian Civil Rights-excluded Economist

• Came second in the Kavel SA by-election 20 May 1992

• Renowned Public Educator

• EXPLAINER OF immediate Reform of approximately 25,000 universities

• the new Neurocognitive Health

• Post post Socratic 2-5 set series quantum-relations logic ex 9th Century CE Iceland, Iraq & IN POINTS OF NON METAPHYSICAL HISTORICAL FACT all original nations

https://www.abc.net.au/news/2026-08-17/data-centres-fast-tracked-under-new-nsw-scheme/107043984?utm_campaign=abc_news_web&utm_content=link&utm_medium=content_shared&utm_source=abc_news_web

Published in honour of the lives and contribution to the entire human race (ex Plato the growly guy STANDING IN THE MARKET @ Athens a helplessly bellicose Slavery & mercantile imports economy City-state circa 400 years Before Common Era and his 193 future nations explicatory mental construct, the macro phenomenon, of neurocognitively most commonly Right Cerebral Quadrisphere RACE MEMORY people who are not silly or demented or under mid teens age and therefore not non screwed non oppressed non disempowered well & well-nourished adult Conceptualisation-capable

🥸

Exponential

View

Al analysis worth reading 164,000+ subscribers

Upgrade

Notes

Posts

Chat

About

• @ Multiple Authors

The state of the

Al economy

5 min. read 0345 • 14

25 Jun 주

e State of

– AI

onomy