Friday Squid Blogging: Squid Found in Light Fixture
Probably a college prank.
As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.
Comments
Clive Robinson • January 3, 2026 2:19 PM
@ Winter,
Sadly for the last fifty or more years the game has been rigged against the ordinary person so “prosperity” can not happen for the ordinary mortal.
But worse… This century the rigging has progressed to the point where the ordinary mortal can not have “peace” either it’s a continuous series of invoked “War on XXX” where a select few decide what XXX is and often it has no reality it’s just a means to another end.
The idea being in part that the 1% of the 1% of the 1% and their families will not get into any kinetic or similar conflict. So that when ordinary mortals are pressed into service to die as cannon fodder, their estate/assets can be acquired for maybe 1 Cent on the Dollar or less to be “rented out” at way more than a Dollar into the future, free from tax and lien.
Ordinary people are fobbed off by being told that it’s a “trickle down economy” which is complete BS, as just a few moments thinking will reveal by simple maths.
It’s cynically called by some a “K shaped economy”,
https://www.npr.org/2025/12/31/nx-s1-5660842/what-is-a-k-shaped-economy
Because there are two trajectories. and just about every one who will ever read this is on the descending limb to rent seeking induced destitution and abject poverty. As was once said at Davos “You will not own anything”.
Which means the K-Shaped Economy is just another “cover it up” as the “real balance” is with tangible assets being transferred from those on the bottom limb to those on the top limb. And like all things in life if you live by emptying the cookie jar at some point it’s going to be empty and you starve and loose both choice and freedom if not life.
The Venezuelan “issue” is because they do not want to buy into the “American Man’s way” dream that Kippling exhorted the Americans to “pick up” over the Philippines back in 1899, and we can see how that did not work out well today.
It’s clear to anyone who can see and think that those boats being sunk are not full of Fent@@@ but Coca@@@. The lie of Fent@@@ was tried on Canada by the “ill wind that blows” Trumpeta and was soundly rebuked.
It’s actually been known since before 2019 that Fent@@@ comes in from Mexico where it is manufactured on mass much as Syria was doing. But it makes a great excuse for starting the death of tens if not hundreds of thousands for grabbing resources very cheaply.
And I’ve explained in the past why there is the demand for Fent@@@, and in short it’s the real side of the “American Dream” hoey-bunkum that can only happen untill everything is bought up by the few by their criminal behaviour then rented out. Thus re-establishing the King, Church, Baron, serfdom three estates model of the Medieval period.
All that said, I do wish most people both peace in their hearts and minds, and peace in their lives to live modestly and without fear and be able to secure sustainable futures for their loved ones.
Whilst I don’t believe in the con-game of deities that support the King Game, I do believe that in general mankind can improve it’s lot and be become better than it was thus share in a better future for all not just the very few. However there is a price that has to be paid by all and that is by taking social responsibility for society not leaving it to what has become a corrupt hierarchy propped up by authoritarian follower guard labour.
Read Cory Doctorow’s 393c talk he gave just a week ago that I gave a link to,
https://pluralistic.net/2026/01/01/39c3/#the-new-coalition
Or you can now listen/watch,
https://craphound.com/news/2026/01/01/the-post-american-internet-39c3-hamburg-dec-28/
Because it describes a simple step that can put a spoke in the wheel of Enshitification that is created by solipsistic managed Corps not just into the ruination of the Internet, but modern life that has become so badly entangled with it and the almost inevitable downward spiral that is following.
There is however a downside in that it will need some form of balkanization, not just of the Internet but technology in general. Look on it as how surgery cuts out that which is bad and growing at the expense of the good, to protect what is good whilst it still can be.
Interestingly it looks like Canada has already realised this and is acting in that direction.
Clive Robonson • January 3, 2026 4:35 PM
@ Rontea, ALL,
With regards,
“True security requires oversight, auditing, and accountability across the entire chain of trust.”
A couple of things,
Firstly you left out the all important by “Skilled workers”.
All the talk of Current AI LLM&ML Systems doing it is compleate nonsense, as I’ve been pointing out for a while now.
These AI systems can not do “skilled” all they can do is “follow skilled” they don’t even learn they simply “shake the can” to fuzz up the product of “skilled workers”.
Secondly as Cory Doctorow recently said about even moderately skilled workers,
“Your Boss Hates you”
With the reasoning being that managers especially at senior levels actually don’t do much other than “makework”. That is they know that if they don’t turn up for work very little changes, and the organisation carries on functioning in their absence. But what they actually know in their hearts and it scares them, is if “skilled workers” don’t carry out their jobs then the organisation stops functioning in some way if not entirely at that point, oft very rapidly there after.
It scares them so much that senior management are absolutely desperate to get need of “skilled workers”.
Which is why as Cory put it they are an easy mark for AI sales persons.
As I’ve pointed out previously the ones who should be scared of AI are those that are senior and other unskilled management, and those who’s jobs are “old professions”. Where memorising facts, rules and following them –effectively bureaucratic jobs– are the jobs the likes of Current AI LLM&ML Systems can do and do well. But only if the inputs are correctly curated and ordered, which is a “skilled worker” job.
It’s why we see “vib coding” failing. Yes the AI can “sausage machine” style “cut code” but it needs a very “skilled worker” to make it not just “secure”, but Safe, reliable and thus available when required as required.
Cory Doctorow makes another point to consider,
“Software is a liability not an asset.”
Which is what “available when required as required” highlights.
Because as Cory goes on to sketch out it’s the actual process carried out that is the asset, that provides the “value added” not which software accomplishes it.
The other issue is of course the old “file format” issue, work is a series of linked chains, as are all supply chains from the micro to macro scopic. You have to be able to efficiently take the output of one software tool and feed it without issue into the next software tool in the chain.
Anyone who had to wrestle with Microsoft and similar closed proprietary software and importantly “file formats” back in the 1990’s knows that the “file formats” were the most insidious form of “Vendor lock-in and forced upgrade”.
Something the Open Source and other Open initiatives started to kill.
As I’ve mentioned on this blog from about Day 1 I always go for “plaintext open file formats” such as CSV and similar. Because they are easy to transcribe to the input of other tools. Or enable me to make other tools.
Back in the early to mid 1990’s “document processing” in more than make, format and print became a thing. There were very expensive proprietary systems that were the forerunner of closed search engines.
All they really were –and still are– was a database of all documents an organisation had made where searches on the database could not just pull up entire documents but link them together via searches on “information within”. This used to be the jobs of file clerks and institutional librarians (jobs that nolonger really exist).
The barrier to making a system was not the data base or the searches but the input file formats… Now consider what happened when I used two backend databases, the first an ordinary relational database, the second and more important what was then called a software “Source Control System” or these days a “Version Control System”.
I designed it initially to take “Postscript” input which back then was a high end print format that just about all “Desktop Publishing” and “Word Processors” supported as well as Fax software (yup it was the 1990’s) it made much of the non productive office work “push of a button” thus productivity jumped way beyond expectation for minimal cost.
Crack the “file format” issue even today and you will get productivity increases… You can then with your own private curated LLM&ML system do the equivalent of what “desktop publishing” used to do but for semantic rather than visual content and style.
But in no way does it replace “skilled workers” it just provides augmentation tools.
zzx • January 3, 2026 6:27 PM
How the defence sector is battling a skills crisis
https://www.bbc.com/news/articles/clyd1lpp1lyo
‘Salary, career path, and job security all looked good, but ultimately a defence sector career “didn’t sit well” with him. “It’s one of those jobs where you don’t want anything you work on to be used.”
That unease about working on lethal technology is just one of many factors contributing to an ongoing skills gap in the defence sector. And that gap could widen as the UK government – like many of its allies – looks to boost defence spending while facing an increasingly volatile geopolitical environment.
Earlier this year, the Ministry of Defence announced it would invest £1bn in AI-powered battlefield systems and announced a new Cyber and Electromagnetic Command. But the military and its suppliers face fierce competition from technology firms and business in general for specialists across these areas.’
Ismar • January 3, 2026 9:03 PM
It looks like there was a cyber component in the cutting of the power in Caracas
“ The lights of Caracas were largely turned off due to a certain expertise that we have”
Clive Robinson • January 4, 2026 5:21 AM
@ ResearcherZero, ALL,
With regards,
“Under international and U.S. law, the Trump administration’s actions, the incursion into Venezuela and the capture and rendition of the Maduros, are blatantly illegal and criminal.”
Serious as that is, it’s not in the list of real issues that concerns me.
The issue top of the list is,
1, Might is right.
That is the belief that if you have “the power” then you can do anything you want. In the “King Game” this is known as “Divine Right” and it’s killed more people than any other human failing known.
As such views are held and acted upon by those who are abnormal mentally you can seriously say that,
“The lunatics are running the asylum.”
And this can only end one way which brings me onto the second bullet point on the list
2, Sets precedent.
This is a failing the US has made and still makes over and over since the end of WWII.
What is to stop another country grabbing Trump or his wife or other members of his family and “hanging him from the yard arm?” By international “common sense” and “collective defence” it was seen that certain actions should not be taken. One of which is that “Heads of State” and their families should not be taken, imprisoned, executed, or assassinated. But the US has repeatedly done such acts for short term political gain and created far more long term problems not just for themselves but others.
3, Dictatorship by Exceptionalism.
The US is increasingly saying that it’s “wants” should be global but nobody else should have the same rights or the right to say no.
It’s a form of solipsism that is rather more than egotism and it is exceptionally harmful, not just to others but US citizens as well.
I could go on with the list but the reality is for short term politics the US Executive have blessed the lunacy of Corporatism. This action has never been about what the US Executive claims, but about unlawfully “grabbing resources” and similar as “pay-off”.
It will not benefit the majority of US citizens as any “gain” will stay “off shore” and actually act against US citizens interests thus be a “clear harm” to them.
I’ve made warnings about this sort of thing since the 1990’s and on this blog since it’s existed. The fact it’s all been “logically predictable” to so few people is actually quite shocking.
Just remember I’ve made predictions about the US fomenting war against China and Iran for nearly as long and if people care to look they will see that the notion of “boiling the frog” applies… And yes we are heading down that road where the destination is kinetic disaster of global proportions.
Do the people of the US want crosses painted on their children and grandchildren’s backs, and for them to be used as “food for the cannons?”
If you remember back the US Executive has been saying that Europe needs to spend 400-500% more on “defence” on the pretext that Russia will invade. It appears reasonable, untill you step back and look at it from the view point that it actually means
“You will give it to the US to build up it’s pre-global conflict MIC development”
That is the US does not have the manufacturing capability currently to “go to significant conflict”. It needs to build up the “industrial” side of the MIC which it can not afford to do. As the Corps have killed “manufacturing” in the US and “off shored” not just the money but skills and other necessary capabilities. So the idea is make other countries pay for the US to get it back first then it can start global conflict.
But also consider that US legislation has made it so that all US Manufactured items must have the equivalent of “kill switches” built in… (It’s most likely why the lights went out in Caracas).
Thus using US weapons for defence is not actually giving “sovereign security” it’s surrender to short term thinking of US political whims and solipsism…
Clive Robinson • January 4, 2026 7:47 AM
@ Ismar, ALL,
You note that,
“It looks like there was a cyber component in the cutting of the power in Caracas”
I’m fairly certain of it from just basic logic.
Power generation is nearly always under control in many parts of it by “Industrial Control Systems”(ICSs) which connect back often by Microsoft Windows based “Remote Telemetry Units”(RTUs) that inturn conect back to a more centralized “Supervisory Control And Data Acquisition”(SCADA) usually Microsoft Windows based system, where the system operators are based.
There is no difference in this architecture to that which was attacked by “stuxnet” all those years ago.
As we know three basic things are almost certainly true,
1, The comms beyween ICS and SCADA was across the Internet (they nearly all are due to bean counters these days).
2, That any “signing certificates” are almost certainly easily compromised.
3, That a “back door” has been sent out as an update from supply chain compromise.
The US has considerable form on all of this from the stuxnet attack.
Likewise Russia has been doing it to rather more than just the Ukrainians.
Very few Nations have considered this a Sovereignty Issue thus taken the basic Cyber-Defence steps you would think “National Security” would demand.
But further consider back in Gulf War I the use of “smart munitions” loaded with conductive carbon filaments, to drop on pylons and switch yards to take out power grids was a weapon in the US arsenal that was “revealed” at the time.
So yes it’s well within “Known US capabilities”. But to tell which method was used can be seen from how long it will take to bring it all back up.
ResearcherZero • January 4, 2026 8:07 AM
@Clive Robinson, ALL
You have in part described the plot of The Manchurian Candidate, or perhaps because I was reading it, I am drawing parallels with tactics of political operatives, that seek only to benefit the few by stooping to any depth to transfer wealth and power away from the public.
This transfer of wealth into a few private hands will increase enormously over the decade.
A group of Wall Street investors took a big bet on regime change by purchasing Venezuelan debt. Pressure exerted by investors on bonds saw a 101% increase in demand on the chance of Maduro’s fall. Investors had lobbied the White House to remove sanctions on Venezuelan debt in the form of bonds. $60 billions in these bonds had been snapped up at very cheap prices.
‘https://www.ft.com/content/a8beec5e-0c3f-4fb7-8780-6eeaa6c0f1ab
The value of bonds holding Venezuelan debt has significantly grown in a very short time.
https://www.reuters.com/world/americas/venezuelas-billions-distressed-debt-who-is-line-collect-2025-12-19/
Better returns gained through diversification of investments outside of the U.S.
https://www.ft.com/content/10a8a099-5719-42ce-a2eb-edc3045a632f
Clive Robonson • January 4, 2026 6:18 PM
@ ResearcherZero, ALL,
Part 1,
With regards,
“A group of Wall Street investors took a big bet on regime change by purchasing Venezuelan debt.”
This is not the first time on South American country Debt. Remember another SA country declared it’s self “bankrupt” and a group of investors,
1, Bought up all the worthless paper.
2, Got a US court to give them all the Dollar holdings of that country in the US Fed Reserver for that paper.
The result the investors became extraordinarily wealthy. The people of the country nearly staved to death (many certainly died of poverty). This is the solipsist way as they view it as being totally OK.
Clive Robinson • January 4, 2026 6:58 PM
Part 6,
Because the honest,
“Of course not…”
Is not going to be admitted to even though simple logic tells why there is not a chance of that happening…
It’s for two basic reasons that are actually the two faces of the same coin,
1, Basic economic supply and demand.
2, Because it’s “too useful”[1]
Even though it kills something approaching 100,000 Americans a year with around half being illegal Fentanyl and similar entirely manufactured and easily produced drugs. The nature of the actual drugs that kills US citizens is not easy to determine after death,
‘https://www.medcentral.com/meds/opioids/fentanyl-separating-fact-fiction
Worse the likes of Fentynal, Fenetylline and similar is they are used as a filler in other “street drugs”… So if you think you are getting one thing, your are actually getting an unknown cocktail of synthetic drugs –made by very similar processes– all of which have varying complications and contra-indications. As the production processes are similar switching from manufacturing one to another is not difficult and thus cross contamination can easily result.
A true fact is the deaths save the US Federal and State Govs save vast amounts of money as those that end up dead do not live long enough to collect on pensions medicare and similar “expensive but obligatory” welfare. Thus allows vast profits to continue to flow into certain pockets.
The legal manufacture of the raw / feed stock of these drugs / fillers is being made in China and increasingly the worlds largest manufacturer of generics India. And legally exported by them not to the US but other nations where they get “lost in the system”…
[1] As I’ve indicated before the real problem is the way the US does “health care” it’s a money machine that provides unimaginable income to certain interests. Especially “pain relief” which should be treated in other ways not by just upping drug usage. But the illegal drug supply is a very useful shield to stop US citizens going after the “Health care crooks” because rather than have those who have been failed by the system killing themselves by suicide or worse they get pushed into the consumption of illegal drugs thus can be “victim blamed”.
[2] The evidence for Syria manufacturing Fenethylline and supplying into the Middle east is fairly conclusive. However their earlier supply of Fentynal to various countries is less direct and found not by “finished product” but by records of imports of raw / feed stock. The argument given by some is that Syria stopped supplying the “Americas” when the Mexicans started to undercut them (basic economics of “distance costs” which always makes local manufacture more profitable thus causes “markets to fracture” which is not what Corp/cartels want, thus getting “state intervention” in various ways). Others indicate the increasing difficulties due to “sanctions” bringing scrutiny on exports. Either way they apparently shifted from one to another.
lurker • January 4, 2026 11:16 PM
“We’ve fixed it, and now you can use Microsoft’s or Google’s authenticator apps”
Meanwhile another local social media site shows its ignorance of hacker culture by asking for a court injunction to prevent the perps doing anything bad with the “stolen” data
Clive Robinson • January 5, 2026 6:22 AM
@ Bruce,
A couple of threads back you asked,
“Are We Ready to Be Governed by Artificial Intelligence?”
And the thread following it,
“Using AI-Generated Images to Get Refunds”
Now consider those in terms of how AI can spoof things… Then ask,
Are we ready to be owned by AI using generated faux-biometrics to get authentication and access?
I’ve warned in the past with all these “think of the children” nonsense biometrics that the systems have gaps in the chain between the meatbag and the decision process. Well I’ve yet to see a way to stop Current AI LLM&ML systems “filling the gap thus rendering what comes before it moot.
Almost the first thing those not with a sufficient technical background will say,
“But this will only be the case for remote biometrics not local biometrics.”
Thus arrive at the notion of using a token / dongle at the user end of the chain that communicates back via a cryptographic or similar “secured” communications path.
But a moments thought reveals that this just makes the gap at the user end where they can more easily exploit it.
Thus within a year or two at the most the “something you are” will be “dead and buried” as an authentication factor.
But think a little further and so to will be “something you have”.
Thus leaving “something you know” but that will fail for another reason…
The human memory is remarkably bad at accuracy it’s the price we payed tens of thousands of years ago for survivability. That is we don’t need to know in detail how a predator looks, to recognise a threat. We just need to recognise they way they behave when they are preparing to attack us.
The problem we have, with Current AI LLM&ML Systems is that they really don’t do “learning” in real time. Thus you can not use AI to defend against an offensive AI system.
We’ve seen this demonstrated with the endless cat and mouse games of “prompt attacks to jail break”.
As I’ve indicated before the LLM “Digital Neural Network”(DNN) is really a DSP “adaptive tuned filter” working with many dimensional layers of spectrums. In the likes of audio DSP systems the adaptive aspect can be made very close to “real time” but due to various issues not so for LLM DNN’s because of the way the ML component works. And if you think about it this is not something that is likely to change, the defending LLM is going to be a very long way behind the curve of the offensive LLM… That as the old adage puts it,
“The attacker only has to win once, the defender every time…”
Ordinarily you could “chain low reliability tests” to improve the odds of detecting an offensive move… But the human mind is so unreliable that such a strategy is going to over load it’s capabilities.
In effect we’ve lost or are about to loose the battle on the three main authentication factors, we just as an industry don’t realise it yet…
Clive Robinson • January 5, 2026 7:54 PM
@ ALL,
Real world AI in the Office, about what you would expect…
UK government trial of M365 Copilot finds no clear productivity boost
“AI tech shows promise writing emails or summarizing meetings. Don’t bother with anything more complex.
A UK government [major] department’s three-month trial of Microsoft’s M365 Copilot has revealed no discernible gain in productivity – speeding up some tasks yet making others slower due to lower quality outputs.“
https://www.theregister.com/2025/09/04/m365_copilot_uk_government/
Remember the little hamsters in their wheels were only a fraction of the Dept staff and were actually not a true random selection (so not a gold standard “double blind” test).
Also this had me thinking,
“And hallucinations? 22 percent of the Department for Business and Trade guinea pigs that responded to the assessors said they did identify hallucinations, 43 percent did not, and 11 percent were unsure.”
Hmm 22% + 43% + 11% = 76%
That’s 1/4 of participants AWOL…
So multiply those percentages by 4/3.
22% becomes 88/3 which is around 3/10ths see hallucinations.
But we also see that MIT statistic rolled out again,
<
blockquote>“An MIT survey published last month, for example, found that 95 percent of companies that had collectively sunk $35-40 billion into generative AI had little to show for it.”
<
blockquote>
Now I’ve mentioned before the known fact that only about 1/10th of “new brand products” based on an existing commodity that make it to the market place “as a trial” actually make it to the point of “making a return”.
So the 1/20th of “completely new products” could be seen as either an “abject failure” or “comparative success” depending on how you look at it.
Either way it’s actually slightly better than you would expect based on other “completely new products”
Which is why these things are not “gold standard” trial results in either case.
Because, in the UK Gov Dept case they were mostly “self selecting” (volunteers). And in the MIT case they were again “self selecting” (first mice).
Clive Robinson • January 5, 2026 8:20 PM
@ ALL,
In the past I’ve pointed out that the “Digital Neural Network”(DNN) in an LLM is actually a very large “Digital Signal Processing”(DSP) filter. And that the ML makes it in effect “adaptable”.
I also mentioned that what it was filtering was “semantic layer spectrums” but did not really say what they were because there are not many easy to comprehend examples out there that are as “real world” as they could be. Also explaining ML as “chunking” to make “tokens/vectors in scope” is all a bit “airy fairy, up in the air, hand wavery”.
Well today I read a couple of articles that were both “real world” and relevant.
The first article is about an issue in the US legal system to explain why searching “case law” and similar is so expensive and actually an unlawful cartel. But in the telling of the story it does give a reasonable example of “semantic layers”,
https://www.thebignewsletter.com/p/gatekeepers-of-law-inside-the-westlaw
The second is about the issues to do with “ML Chunking” and highlights just some of what can be involved with getting it right,
https://minha.sh/posts/so,-you-want-to-chunk-really-fast
If you read them both then go back to the first again it makes the second even more clear.
ResearcherZero • January 5, 2026 11:48 PM
Binary classification systems used by algorithms set different parties in opposing and competing positions, further amplifying disagreement by incentivizing these divisions.
‘https://thebulletin.org/premium/2025-12/division-begets-division-in-the-age-of-algorithmic-classification/
As treaties expire, it’s worth considering nuclear winter could push Earth far from habitable for humans and animals. Previous studies have underestimated how great species loss and the rapid reduction in food production would be.
World leaders have failed to grasp that the majority of all species – along with food sources – could become extinct. Land, sea and food source species humans need to survive
Compounding effects of warming, pollution, deforestation and nuclear winter.
https://www.cell.com/heliyon/fulltext/S2405-8440(23)02428-3
The indicators for nuclear deterrence are all heading in the wrong direction.
https://thebulletin.org/premium/2025-12/eighty-years-and-89-seconds-its-time-to-fight-against-midnight/
ResearcherZero • January 5, 2026 11:58 PM
The effect of algorithmic driven decisions may already be negatively impacting policy choices.
Posturing and nuclear competition will increase in Europe over the next decade.
‘https://thebulletin.org/premium/2025-12/the-changing-nuclear-landscape-in-europe
The Monroe Doctrine originally championed the independence of Latin American countries. Instead it led to repeated U.S. military intervention, power vacuums and instability. Many of the security threats to the United States did not begin in the Western Hemisphere and will continue to arise in other regions. Problems such as the drug trade, mass migration of people fleeing conflict or failed states and terrorist groups, emerged from imperialist intervention.
https://historyhowithappened.com/how-the-monroe-doctrine-turned-into-american-imperialism/
“If the future resembles the past, the result of our endeavors will not be freedom but the kind of anarchy in which extremists …can flourish.”
American foreign policy promoting regime change has repeatedly led to failed states.
https://lawliberty.org/the-use-and-abuse-of-american-foreign-policy-doctrines/
Combined with cuts to foreign aid, preparedness and advanced planning to respond to events that threaten global health, exacerbate humanitarian crises and destabilize fragile regions – along with insight and surveillance of other factors that could undermine global security – has been seriously degraded.
lurker • January 6, 2026 12:15 AM
@ResearcherZero
Those who do not learn from history are condemned, period.
Local paper has a cartoon, unfortunately behind payawall for web; suffice to say it is an excellent illustration for the words of W.H.Auden:
Clive Robonson • January 6, 2026 7:30 AM
@ ResearcherZero,
With regards the “Doomsday Clock” Xmas/New year message, that you link to
Unfortunately the author Ms. Bell comes across initially as naive then fails to go into psychology of authoritarianism in any depth and totally ignores the economics and how deterrence is not a zero sum game.
Also Ms. Bell shows lack of credibility as many nuclear scientists do with regards the nature of “existentialism”.
OK Ms. Bell was once a diplomat but the public want punch not passivity, that is they want News Style not Conference or Talks style.
But worse she also claims a drop in the number of “physics packages” from over 70,000 to 13,000 or less as a success. Conveniently ignoring the fact that dropping to 1/5th the number does not in any way make up for the fact the yield has gone from kilotons to MTonnes with the square root[1] of 1000 being a little under 32. Though it should be 10,000 to 50,000,000 thus sqr(5000) or 70 increase in coverage (with supprisingly less fall out).
The problem is that authoritarians want more than power, they want eternal remembrance as well… So want to “go down in history” for a “destiny”, and care not a jot about how they get it… With history showing in general we don’t remember “good” but we very much remember “bad” you get to see why some people do what they do.
Ask yourself what the difference in mentality is between a nut-job driving into a crowd and then chopping at people with what are machetes or similar and someone “pushing the button”.
If I mention “Richard Reid”… would you remember him as the “failed shoe bomber of very low intelligence” or a “dangerous nut-job allegedly intent on mass murder”? Does it matter to him as long as his name is remembered?
What of Reid’s two co-conspirators?
Did you even know there were any and that one was previously a professional football player?
And that they are now both free to walk in society?
Now consider “The British Dentist” and now exiled leader of Syria wanted on “war crimes” charges and living in a flat in Russia. How about the now dead former leaders of Lybia and Iraq and probably soon to be dead leader of Venezuela?
What about the current leaders of the three world super powers, which do you think will be remembered the longest?
Even William Shakespeare who lived over four centuries ago clearly understood the idea of authoritarian leaders and their desire to be both autocratic and eternally remembered…
I could go one but hopefully I’ve made the point about authoritarian leaders.
As for economics, the real cost of nuclear weapons is immense because you not only have to have delivery mechanisms, you have to defend the entire supply chain of both from attack both internal and external.
It’s why both South Africa and Ukraine gladly gave up theirs. But what did they get for doing so?
Now consider India and Pakistan. Both were treated as of no worth prior to becoming nuclear, now they are nuclear the have seats at the top table… Though their economies are markedly different. Whilst India’s leaders name is currently well known, can you name Pakistan’s unless “fresh in the MSM” you’ve read. How about the name of Pakistan’s “Father of their Bomb”?
Who incidentally gave the bomb secrets not only to Pakistan, but Libya, Iran, North Korea and it is said Iraq as well. The leaders of Libia and it’s assumed Iraq decided the cost was to high thus did not progress, they are both now dead and their countries now corruptly lead vassals of an “Empire that claims it is not” that is stripping them of their resources and encouraging “terrorism” in a “Cecil Rhodes” style plan.
The issue of authoritarian leaders and economics are something I think needs to be top and center if you want to ever hope of getting rid of nuclear weapons…
[1] It is actually quite hard to gauge the yield of nukes which also means maximum blast wave radius is awkward to determine at best. As a first approximation the device has spherical thermal expansion which transitions from volume (cube root) to area (square root) for the blast wave. The use of area for scaling damage radius is one used for rough approximation, but volume is used for arguments in the number of weapons in Multiple device delivery systems. So pick your horse either way and hope you never have to ride the race.
JG5 • January 6, 2026 7:43 AM
A version without the links that triggered moderation.
The “health-care” topic slipped by before I could say anything useful. That is yet-another symptom of failed government. I picked up a very important clue in this blog, at the intersection of Kaiser and Nixon, which was captured on the tapes. That is going to be front and center in the debate over electricity and other cost of living issues in the US.
For reasons that escape me, I remembered a discussion with MarkH of fires in California and the swimming pool escape route. I think that fire safety has been discussed politely many times.
“Those who do not learn from history are condemned, period.” – lurker
Indeed, there was a nightclub fire in Rhode Island in 2003. Involving flammable foam material behind the stage, hit by fireworks.
There was an unfortunate in-flight fire in 1968. Involving flammable foam material on a cot/bed/bunk, hit by bleed air. If you are going to crash your B-52, you should leave the four 50-megaton hydrogen bombs back at the base.
I think that @Clive had comments on Grenfall at the time. I was mentally alert, but may have felt too far removed to have much to say. More foam material hit by a positive source of ignition. The Medieval UK might have provided more fitting punishments for the perpetrators.
Much more recently, in what otherwise seems to be a safe, clean, modern country, a repeat of 2003. Wherein fireworks hit some foam sound-damping material in a nightclub.
I know that wikipedia is ground zero of the information wars, but most of the articles that I see are accurate. I am sure that other sectors are more polluted. I was too lazy to see how many of the fires involved foam materials.
Clive Robinson • January 6, 2026 9:46 AM
@ JG5
Re
“… how many of the fires involved foam materials.”
They are often not called “foam” or even cloth these days but “low density” or “high thermal insulating” or similar with an attached “R-Value”.
Thus the “R-Value” indirectly gives an approximation / indication of how lethal it is going to be to you…
I was actually shocked “by the news” of the revelers in that basement bar “Le Constellation” fire in the Swiss ski resort just a few days back. The report indicated they were more intent on getting footage on their mobile phones than hot-footing it out of there whilst they still could… If true or not it’s still a shocking thing to say. It’s been indicated that there are over 115 seriously injures some of whom who will be sent to other countries for treatment, and that one of the 40 dead is a 14year old local girl.
But more inexplicable and may be cause of the news report is that some of the dead actually took video of the fire with one video of the only set of stairs starting to burn then getting engulfed as others panicked trying to get up them.
It’s being reported that the owner who was from Corsica and had served time in prison, had halved the width of the only stairs out in DIY renovations made with wooden slats and acoustic materials in the ceiling not designed to be used that way. And that his wife was related to senior fire officials in the area and for some reason the bar had been very infrequently inspected for fire safety, and might not even have been fully inspected after the renovations.
Sometimes “your security” is best served by “basic situational awareness” and consideration of what “could happen”. As my father used to point out,
“The place to be when there is trouble is somewhere else.”
With the observation that using your eyes, ears, brain, and physical positioning would assist in that task by giving you both a “heads up” and importantly “time to act” and “get the heck out of Dodge”.
Clive Robinson • January 6, 2026 11:16 AM
@ ALL,
This may be something or nothing.
It all depends on if what was going on can be “tracked back” etc.
News: There Were BGP Anomalies During The Venezuela Blackout
“When watching the situation in Venezuela unfold, the phrase “It was dark, the lights of Caracas were largely turned off due to a certain expertise that we have” caught my attention
…
BGP is the first thing that comes to mind. It’s a protocol used by routers to determine what path data takes to get to it’s destination, it does this by exchanging routing information between Autonomous Systems. It is also notoriously insecure and much of the data about BGP is collected in public datasets.“
https://loworbitsecurity.com/radar/radar16/
Have a read and make your own mind up.
But it is odd this security incident has not been highlighted more generally.
Clive Robinson • January 6, 2026 5:12 PM
@ Bruce, ALL,
You will love this new bit of law…
I wish the UK had the equivalent “send to bin or hell now” buttons for adds.
I use Brave with javascript disabled so rarely see adds but the few I do I’d rather not see at all. Also I regard the bandwidth taken from me as “theft” so I see online adds as “criminal activity” regardless.
Yes I know a lot of peoples jobs and income “supposedly” are based on “add income”… but really, lets be honest now, most are actually being scammed by their employers just as much as the people who pay for the online ads in the first place are being scammed… The whole bit in between the company buying ads time and consumer eye balls is as crooked as heck and beyond. And would be regarded as a form of fraud if not for those oh so crooked lobbyists.
Clive Robinson • January 6, 2026 6:00 PM
@ ALL,
Fun for those that can think backwards.
This “data diode design”,
https://nelop.com/bespoke-data-diode-airgap/
Cane to my attention via,
https://news.ycombinator.com/item?id=46516117
Before you read this Ycomb page, see if you can work out two things,
1, Why it’s not even technically “air gapped”.
2, Why “air gapping” is not a good idea these days as “energy gapping” is preferable.
For those that are long term readers of this blog, will know that this design is a rehash of one much discussed on this blog several years ago.
And that I’ve given the answers to my two questions to @figureitout around the time of BadBIOS.
P.S. @figureitout and the rest of “the usual suspects” if you are still lurking pop up and say hi.
Clive Robinson • January 6, 2026 10:56 PM
@ Bruce, ALL,
And so it begins, in a factory far far away..
Looks like those pesky LLM&ML systems are going to be getting “agency this year. Wether it’s “free” agency or not could be a major hurdle,
Optimus Schmoptimus – Boston Dynamics’ humanoid robot is already in mass production
“Remember when Elon Musk predicted that there would be thousands of Optimus robots at Tesla factories by the end of 2025? Well, that didn’t happen, but competitor Boston Dynamics has just announced that its humanoid robot, Atlas, is going to the big time.
…
In addition to its planned Hyundai deployments this year, Boston Dynamics also announced a partnership with Google DeepMind at CES that will see the pair working out how to integrate Gemini Robotics AI foundation models into Atlas to, according to Boston Dynamics, “give the robot greater cognitive capabilities.”“
https://www.theregister.com/2026/01/06/boston_dynamics_atlas_production/
Wether to be delighted or horrified is something I guess only hindsight will determine…
Fun point some will remember that I said “giving AI ‘free’ Agency” to explore and learn about it’s environment was a prerequisite for the I in AI to actually having meaning in the meatbag sense.
So is the limb I’ve climbed out on looking shaky and possibly going to leave me “with egg on my face”?
Honestly I don’t know, but it’s a prediction I’m going to stick with on one condition,
1, They first fix the LLM memory problem.
It’s been shown that LLMs do not actually learn in the “long term memory sense” as they interact with a user or other external stimulus. That is the weights and biases in the DNN do not dynamically change.
So Current AI LLMs do not “learn by doing”. I’d kind of assumed it was a “relatively easy fix” issue they would have fixed effectively if not efficiently by now. But they have not so far, which technically it does not appear on the face of it to be difficult to do, just potentially time and money consuming, so “slow and expensive”.
So on the assumption they fix that issue fairly quickly, then personally I’m “cautiously curious” and,
“Wondering if they will need a “double tap to the head” like zombies, to stop them chowing down on you” 😉
Any way it’s a “crystal ball” notion and I could be shown to be wrong that “agency” is a key missing step to “actual I in AI”.
In which case you guys are going to have fun kicking my ass (all those jokes and so little time 😉
But seriously I’ve presented my reasoning, in the past… does anybody think I’m wrong about “Giving AI Agency” is a prerequisite for biological level Intelligence?
lurker • January 7, 2026 1:07 AM
@Clive Robinson, ALL
re giving AI “agency”
When humans start learning they are small and fragile. Their physical and intellectual environments are carefully guarded by their carers. I don’t expect the techbros to have much inkling about similar protections on their precious machines. Free rein learning is rare in humans, and with mixed results.
I recently used a reverse image lookup to identify a detail of artwork embedded in a music file. I took a selection of search engines offered by DuckduckGo. Three claimed to be AI, or AI asssisted. Those identified the art as: sawn timber, gravel, or an RGB vertical test pattern. Tineye showed me 40 examples of the CD cover, unidentified, and two posts from Tumblr which correctly identified the art. I don’t know how Tineye does it but they are the best I have found in my use cases.
Someone who knows about this told me it was likely the amount of skin tone triggered a content warning threshold, and the AIs deliberately lied. Also even the art specialist AIs are very poor at this work because i) so little historic fine art has been properly digitised and published on the ‘net; and
ii) most of the supporting literature is is non-english languages.
Put autonomous robots with flying cars for the meantime.
Proxy voting with AI
“The [JPMorgan] unit, among the world’s largest investment firms with more than $7 trillion in client assets, has to vote shares in thousands of companies.
This coming proxy season, it will start using an internal artificial-intelligence-powered platform it is calling Proxy IQ to assist on U.S. company votes, according to a memo seen by The Wall Street Journal.
The bank will use the platform to manage the votes and the AI also will analyze data from more than 3,000 annual company meetings and provide recommendations to the portfolio managers, the memo said, replacing the typical roles of proxy advisers.
JPMorgan thinks it is the first large investment firm to entirely stop using external proxy advisers, which provide much of the industry’s plumbing, the memo said. […]
JPMorgan Chief Executive Jamie Dimon has been one of the most outspoken critics, telling an industry gathering last spring that proxy advisers are “incompetent” and “should be gone and dead, done with.””
JG5 • January 7, 2026 3:44 PM
@Clive – Yes, it was that Jacques Baud. Perhaps Freedom of Expression is a better fit in this squid topic. The European Charter appears to guarantee freedom of thought and freedom of expression. Very similar to US attacks on speech perpetrated under Biden.
https://www.schneier.com/blog/archives/2025/12/denmark-accuses-russia-of-conducting-two-cyberattacks.html/#comment-451057
@ResearcherZero – Also a good place to follow-up the attack on Putin’s residence, by parties as- yet unknown. I thought we might infer some information from the response.
https://www.schneier.com/blog/archives/2025/12/denmark-accuses-russia-of-conducting-two-cyberattacks.html/#comment-451041
The Russians may have sent one or more messages at times and places of their choosing. The main reason that I moved it out of the cyberattack thread is that drones don’t fit the classic cyberattack profile – a clean definition might be “sending a signal or combination of signals that causes the enemies/adversaries hardware to do something against interest.” That is a form of “projected intent,” as I defined it many years ago. A drone attack also uses or is “projected intent,” but the code runs on the attacker’s hardware. The tibia in 2001: A Space Odyssey also projects intent, but on much shorter scales of time and distance. Without actual autonomy. Hypersonic missiles are a new twist in the game of projecting intent.
Did Russia Just Send a Message to Donald Trump? Who Ratted out Maduro?
https://larrycjohnson.substack.com/p/did-russia-just-send-a-message-to
Larry C Johnson Jan 07, 2026
…
On January 6, Russia launched three major missile strikes against US-owned facilities in Ukraine.
…
I don’t know if this fulfills Russia’s promise to retaliate for the US-backed December 28 failed drone attack on Putin’s official residence in Novgorod, but I don’t think it a coincidence that three US-owned facilities that appear to produce or warehouse weapons for the Ukrainian war effort were attacked on the same day.
…
Clive Robinson • January 8, 2026 7:00 AM
@ Bruce, ALL,
European Capital power hit by claimed EcoTerorism
I’ve seen little of this in MSM outside of German reporting.
But what have been described as Left-wing Eco Terrorists have claimed they were responsible for destroying power cables out of a Gas Powered power station in Berlin, plunging something like 50,000 busineses, homes, education, hospital and emergancy service centers into not just winter darkness but well below zero temperatures.
The incident has also critically effected emergency services and other infrastructure causing flooding by raw sewerage backup into peoples homes etc. As well as inability to carry out maintenance / repair on emergency vehicles etc.
See more,
German prosecutors open terror probe into Berlin blackout
“German federal prosecutors on Tuesday said they had launched a terrorism investigation into an arson attack on high-voltage cables that triggered a power blackout affecting about 45,000 households in Berlin.
Prosecutors said they were probing Saturday’s attack on suspicion of “membership in a terrorist organization, sabotage, arson and disruption of public services.”
…
The attack was claimed online by a far-left extremist group calling itself Vulkangruppe, or Volcano Group, which said it was targeting “the fossil fuel economy” driving climate change.“
https://www.dw.com/en/german-prosecutors-open-terror-probe-into-berlin-blackout/a-75413616
Put simply a major design weakness was probably exploited that is a major security vulnerability found in many places in the world.
The weakness was that a bridge with public access had the main power station output cables “slung underneath” with little or no protection.
Thus all that was required was a fairly minor fire to destroy the cables.
The resulting blackout has been called the worst since WW II
Interestingly London UK learned the hard way from WWII that power cables that need to cross rivers, need to be in deep tunnels underground. Also from PIRA terrorism last Century that all access points to sub surface major infrastructure need properly secured access.
Unfortunately it took hundreds of “Cable theft by Land Rovers” to make people realise that even minor sub surface infrastructure needs access protection, something that is only very slowly progressing in the UK Capital.
As far as I’m aware “arson” has not been verified, so as with a major fire on the London Underground I have good cause to remember it might have been build up of wind blown or similar rubbish that then became ignited in some non deliberate way.
Also the fact a claim is made “supposedly by an extremist group” does not mean it came from the group, or that the group actually exists, or actually carried out an attack.
It’s very much in vogue at the moment for false claims to be made from the “supposed” most Powerful man in the world, the US President, all the way down to low intelligence shunned teens wanting to be noticed etc. Such is the state of societies currently.
fib • January 8, 2026 9:06 AM
@ResearcherZero, lurker,
Sometimes it seems that those who do learn from history repeat it on purpose.
Regards
Clive Robinson • January 8, 2026 10:44 AM
@ fib, ResearcherZero, lurker,
With regards,
“Sometimes it seems that those who do learn from history repeat it on purpose.”
There is a saying in the UK of,
“There’s only one way to crack a nut!”
Whilst it’s true, you have to apply sufficient pressure. And there are several ways to do that…
Of which the traditional smashing it with a rock as a primative hammer is the most uncontrolled, damaging, and wasteful way… It’s a way for all humans and quite a few primates to “put in the effort”…
When thinking about it, it always takes me back to a time when Arthur C. Clark and Stanley Kubrick stood and talked about the opening scene of 2001… Where they had primates kill their opposing tribe using the same principle. That of the hammering process on their skulls… To the tune of a waltz.
Brutal, thugish, bloody, and leathal, with that little rush and dance of victory and throwing the stick up in the air.
Whilst our technology has advanced many thousand fold since the mid 1960’s… our base nature has not changed for maybe 35-50 thousand years. And we pay for it every day.
My father used to wrly observe,
“The wheel goes around, the rut gets deeper, but each revolution still moves ahead all be it on a slightly different course.”
And that’s the problem… It’s the same issue we see with Current AI, things appear to change at the surface. But deep down it’s all just the same. No learning no moving forward, the same issues play out the same again and again. A form of recurrent madness or nightmare.
To quote a P.F.Slone song,
“You may leave here for four days in space,
But when you return it’s the same old place.
The pounding of the drum,
The pride and disgrace
Hate your next door neighbour,
But don’t forget to say grace.
And you tell me over and over and over again my friend, You don’t believe…”
And so “The Eve of Destruction” ever present plays out to the end.
What we realy need is actual humans, not narcissist politicians that don’t learn, and that just want to bicker and play juvenile one upmanship games, whilst strutting back and forth, crowing with less brains than a rooster in a coop, likewise destined for the chop.
Winter • January 8, 2026 12:58 PM
@Clive
But what have been described as Left-wing Eco Terrorists have claimed they were responsible for destroying power cables
People might have become cautious with blaming the left as an earlier case of Eco Terrorism proved to have been a paid contract job from Moscow.[1] None of the saboteurs have yet been identified, so it is still unclear who is behind the declaration, with several linguists claiming the original text was conceived in Russian or by Russians.[2]
But journalists also point to the many “single points of failure” in the utilities networks. This should not have been so easy.
[1] ‘https://www.politico.eu/article/germany-hit-by-suspected-russia-backed-sabotage-campaign/
lurker • January 8, 2026 2:41 PM
@Clive Robinson
re burning bridges/cables
Too frequently I have observed exposed cables in public places, usually telecomms, but sometimes power. My first thought was “axes, fire.” I rapidy suppressed any urge to warn those responsible, because there is a widespread infection of “shoot the messenger”. Anyone capable of thinking that was possible must also be capable of doing it, therefore a terrorist.
Clive Robinson • January 8, 2026 6:56 PM
@ lurker,
With regards,
“Anyone capable of thinking that was possible must also be capable of doing it, therefore a terrorist.”
It’s actually worse than your words might portray to the average person.
You are now seen as a Terrorist simply because,
1, You can think.
2, You can see what others don’t
There is a part of the brain deep inside that supposadly gives you the Ahh-Ha moment of “original thought” to you. Such that say you recognise a cat under a bush in the garden or similar.
But what if the Ahh-Ha moment realy is an “original thought” to the world?
People have been imprisoned or executed for having those and to a certain type of politician (hint hint no names)… You are a wild card, a nail that sticks up thus are automatically a terrorist, unless you have a “licence to research / think” such as a PhD…
I’m cursed in that as our host @Bruce used to say “thinking hinky” is something I do all the time and Ahh-Ha moments for me are like hail stones in a storm, numerous and unfortunately dangerous…
The danger being saying anything even here…
Some years ago Prof Ross J. Anderson kind of “pulled a chair out” in that an invite to a chat about going and doing Doctoral work was being offered up via one of the other Post Docs in the lab… For what now looks very silly reasons I did not take the offer up. If I had I’d have got “a licence to think and speak”…
As you know I still research quietly in my dead tree cave, but apart from an occasional nod toward it, I don’t speak much about it any longer.
These are very much dangerous times where “cancel culture” now has a very dangerous edge to it. It’s no longer those portrayed as blue / purple haired loony left types who “don’t wash” etc. But those with an orange hue from a faux-tan bottle and a desire to cause hurt/harm for fun, pleasure and personal gain through proxy guard labour and the like.
In the UK it is now very much,
“Do as you are told or be rights stripped and thrown away…”
Clive Robinson • January 9, 2026 7:41 AM
@ ALL,
AI Obsesed toilets at CES 2026
It’s some years since I was last at the Consumer Electronics Show in Vegas. But even in earlier times it was “fad obsessed” every year. The only hard part some years was working out “Which Fad”. There were ways you could tell mostly by finding the consumer product that should be the least appropriate to have electronics technology in it…
Apparently this year is no different with AI infested toilets talking to the cloud about well I guess your movements, or lack there of… If this reporting is factual, and trust me it very probably is.
<
blockquote>Tech that helps people outshone overhyped AI at CES 2026
Nobody really needs an AI toothbrush that sends their gums to the cloud.
“[T]his year vendors scrawled “AI-enabled” on all the kit they hope will find its way into your home – while airbrushing away its immaturity and downsides.
Attendees could therefore hear vendors spruik AI toothbrushes and AI toilets and promise to turn snapshots of your gums into health-enhancing insights – just don’t ask about privacy – and a plethora of apps that will use machines to better manage your life – and your elderly parents’ affairs, too.“
https://www.theregister.com/2026/01/09/ai_sideshow_ces_2026/
Maybe I’m getting old and jaded, or my tastes have become more rarefied with age, but it really sounds quite gruesome if not tedious like a teenage fright fest crashing into a wedding reception.
But the question arisses as to who it was really for this year?
A clue comes from
“Jensen Huang’s opening day keynote – were completely consumed by [AI projections]. After a 15-second pre-roll of some eye-popping video game scenes, NVIDIA’s CEO spoke for nearly two hours, said nothing about gaming, launched Vera Rubin silicon for AI training and inference, and framed it all as more relevant to financial analysts than the assembled press.”
Which in some ways smacks of desperation of having traversed the crest flat top, before the inevitable down slope and potential avalanche of doom.
As for what the journalist thought worthy of praise, it was the very very few products that might actually help those less fortunate,
“Those four products show us that tech can be more than just privacy invading, soul destroying, and enshittified experiences. We have amazing capabilities, and when they meet the right motives, beautiful things can happen. That’s the reason I keep coming back to CES: amidst all the dross and waste, there are always bright sparks doing remarkable things.”
I was once one of those “bright sparks” doing “remarkable things”[1] in robotics and AI back in the 1980’s it felt good to be alive back then…
[1] The journalist indicates that the robots on show at this CES had a common problem,
“ready for service as soon as they work out a few lingering issues, such as software, safety, and oh yeah, being able to grip things.
As I’ve mentioned before I’d worked out how to solve the “grip” issue back in the 1980’s after nearly having my head removed by a Puma in a lab that is now part of a University doing doctoral research work. It’s actually simple enough you just “reframe the issue” and the answer “drops out”. Leaving the real issue being making the sensors… But I also fixed the then software safety by combining an “Expert System” with “Fuzzy logic”, but then you could say I was motivated to keep my head on my shoulders 😉
Cephalopod camouflage science
“We want to be able to control this with neural networks – basically an AI-based system – that could compare the skin and its background, then automatically modulate it to match in real time, without human intervention,” Doshi said.
“By dynamically controlling the thickness and topography of a polymer film, you can realize a very large variety of beautiful colors and textures,” said Mark Brongersma, a professor of materials science and engineering and a senior author on the paper.
grima Squeakersen • January 9, 2026 4:37 PM
@rontea re: “without validation and active management, it is simply wishful thinking.”
I had a 38 year IT career with 6 employers (8 if you count buy-outs), the last 35 years of which had ever-increasing InfoSec responsibilities (ending as the highest ranked non-C-level IT employee in an $8 BLN international corporation). My conclusion resulting from that experience is that the vast majority of C-suite executives are looking for exactly what you complain about: security theater. They want to be able to report, without outright lying, that they have taken adequate measures to plausibly ensure the security of the organization, and simultaneously report that they haven’t spent an inordinate amount of money doing so. The easiest way to hide that contradiction (for that is exactly what it is) is to farm the security out to a contractor with experience in making the right noises to provide confidence that they are dotting all the necessary “i”s and crossing all the necessary “t”s. The fact that they aren’t actually doing what they pretend only becomes an issue if and when the SHTF (if then). The fact that the real risk is being borne by investors and employees who have neither a complete view of the situation nor any power to change it doesn’t phase those parties in the slightest.
Clive Robinson • January 9, 2026 10:35 PM
@ Winter, lurker, ALL,
With regards,
“People might have become cautious with blaming the left as an earlier case of Eco Terrorism proved to have been a paid contract job from Moscow.[1]”
There are similar cases where it’s said to be a,
“paid contract job from Washington”
Which we used to say amongst other things were “red flag operations”. Which the changing nature of “intelligence analysis” by state sponsored agencies/entities, guard labour detectives, journalists etc made easy for opponents.
Look on it as “proxy wars” that have been moved from the kinetic physical to informational.
I’ve just caught one such where a journalist checking their sources made the mistake of reading Wikipedia for Citations but not reading what was being said on Wikipedia did not relate to what was said on Wikipedia,
If you see me use “supposedly” or “apparently” treat what comes next as “not able to verify to my satisfaction”.
Yes I make mistakes but every time I see something “new” I usually find,
1, It passes “possible by laws of nature” test.
2, It’s not actually “new” that is there is credible background to arrive at what is claimed.
The problem with that is it’s a new variation on the “Defense Spending” issue. That is neither of those tests are actually evidence/proof that what is “new is true”.
It’s also an underlying issue that also gives us in part “AI Slop”.
What prevents it is “further checking and reasoning”, but at some point it looks like,
1, Procrastination
2, Insufficient primary source
And it’s this that allows “informational proxy wars” to happen.
The usual “Big Example” I give of this was what the originators of “stuxnet” were actually upto. Nearly everyone went with the Washington “no name sources” where as I knew sufficient at the time to reason the actual target from Washington’s perspective was “The Hermit Kingdom” by it’s relationship with Iran and their effective “technology trade” being the “only way in”. The use of Iran as a stepping stone was just whipped cream holding that cherry in place and for getting buy-in from the other partner and as importantly getting “plausible deniability” which worked even after “The Hermit Kingdom” called on Washington’s play.
Though I’m still suspicious of it and lable the whole thing under “apparently” as evidence even now is still lacking.
My father used to put a twist on the old sayings of,
“You can’t make an omelet without cracking eggs”
By combining it with the other well known saw of,
“They say the proof is in the eating of the pudding”
And indicating that even that might not be true by adding a rider of,
“Even if the original intent was to make a hair on the dog, egg nog.”
So always keep in mind,
“They say, you must crack eggs to make an omelet, whilst true the also say that the proof is in the eating of the pudding… But remember the original intent might have been a hair on the dog egg nog to ease an aching head.”
It cautions that you cannot go back from “observed effect” to an “assumed cause”. That it’s not science or facts, just assumptions.
A point I use a lot about “forensic science” and how it’s used to “prove guilt” falsely in Court. Where once the oath of,
“… the truth, the whole truth, and nothing but the truth, so help me God.”
Had meaning.
Subscribe to comments on this entry
Leave a comment
← Flock Exposes Its AI-Enabled Surveillance Cameras Telegram Hosting World’s Largest Darknet Market →
Sidebar photo of Bruce Schneier by Joe MacInnis.
Rontea • January 3, 2026 1:31 PM
Security by delegation is not security. Outsourcing responsibility to vendors or third parties doesn’t eliminate risk—it just moves it to a place you can’t see as clearly. When organizations rely on assumptions that their partners, suppliers, or cloud providers are secure without independent verification, they effectively introduce blind spots into their risk model.
True security requires oversight, auditing, and accountability across the entire chain of trust. Delegation can be part of the operational model, but without validation and active management, it is simply wishful thinking.