I hadn’t known that the NGC 1068 galaxy is nicknamed the “Squid Galaxy.” It is, and it’s spewing neutrinos without the usual accompanying gamma rays.
As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.
ResearcherZero • May 29, 2025 10:58 PM
ConnectWise says a small number of customers were targeted by a nation state attack.
ConnectWise had stated ASP.NET machine keys were used in ViewState code injection attacks.
‘https://www.connectwise.com/company/trust/advisories
ViewState allows state values to be preserved across page postbacks.
Microsoft identified that more than 3,000 machine keys had been publicly disclosed online.
After achieving remote code execution, attackers had set up Godzilla to control servers.
https://www.microsoft.com/en-us/security/blog/2025/02/06/code-injection-attacks-using-publicly-disclosed-asp-net-machine-keys/
Clive Robinson • May 30, 2025 5:44 AM
@ ResearcherZero, Bruce, ALL,
With regards the link to Greynoise labs you give…
The opening paragraph is, –independent of the vulnerability the article is about,– of interest,
“Using an AI powered network traffic analysis tool we built called SIFT, GreyNoise has caught multiple anomalous network payloads with zero-effort that are attempting to disable…”
Using a variant of a ‘current AI LLM and ML system’ to “scan logs” looking for “low level anomalous signals” is something I’ve been thinking about.
As anyone who has tried it knows reading in-depth logs is important, but it’s “drudge work”. Something most humans can not do if even for quite short periods of time.
Yes you can learn to read them like a book, but you can also go mad trying, or both… So it falls to the “chicken or the egg” question.
Conventional automated scanning looks for “Known, Knowns” or sometimes “Unknown, Knowns” but generally not “Unknown, Unknowns”. But whilst they filter a lot of junk thus making the task less arduous, they do throw up a lot of “grass” of “false detects”. Due to this they are in some respects easy to get past due to the back-end human not seeing the “valid signature in the noise”.
As we know LLM&ML systems are just “adaptive filters” working on “tokenised strings of text” and finding correlations to “train the weights”. In effect a statistical tool that can very effectively parse the lines of text from log files.
The trick is getting it to recognise “Low Probability of Intercept”(LPI) signals. This is something certain “Electronic Warfare”(EW) tools currently do via thresholding techniques (and more complex tricks).
Put overly simply you remove the deterministic “know signals”, and events that have very low or no repeated entries of the “random signals” as “out of band” then to use the Martian quote,
“You science the shit out of it”
On what remains “in band” in the middle between what you assume is “deterministic” and “random” that is effectively “the chaos zone”.
You can do this by slowly moving the thresholds you can tune the filter not just to the “regular events” but also the “determanisticaly irregular events” made by adding pseudo random or actual random noise to hide the
probe signal. Basically to search for “hiding patterns”.
Having “pre-filtered” you can then use the AI to do the equivalent of “semantic searches” on higher level meanings to drill down to essential elements that can be investigated in other ways.
There are other tricks and you can find them in books etc on EW and LPI communications systems likewise radar processing in the presence of jamming signals[1]. The idea being to either “select or reject” a signal from other signals, so broadly “select LPI” and “reject jamming”
Knowing how some of such systems work from published works you can then do the classical “Renaissance-Man” trick of taking know solutions or methods from one knowledge domain and apply it to a new knowledge domain.
Like “statistical mechanics” it does not require any real intelligence to perform the drudge work, you just “go through the motions repeatedly” in a way not to dissimilar to “Weather forecasting”. Something the current AI and ML systems can do, but most humans can not.
[1] A classic “amateur” example goes back to the days of the Russian “Woodpecker” Broadband HF Over The Horizon Radar –based just next to Chernobyl– that jammed “Ham bands” from the mid 1970’s through to the end of the 80’s,
https://en.wikipedia.org/wiki/Duga_radar
Basically the signal processing was based on fast “noise gates” that cut the signal path for the high energy radar pulses leaving most of the desired signal “that the operators brain” the filled in the gaps made by the gates.
Clive Robinson • May 30, 2025 6:55 AM
@ Bruce,
You mention,
“[T]he “Squid Galaxy.” It is, and it’s spewing neutrinos without the usual accompanying gamma rays.”
The article you link to notes it’s at quite a varience to the current “standard model” which has a lot of “Dark Matter” “Dark Energy” fiddles to try to keep the model alive in the face of new “conflicting observations”.
As the article notes,
“Now, researchers from institutions including UCLA, Osaka University, and Japan’s Kavli Institute for the Physics and Mathematics of the Universe believe the Squid Galaxy may be producing neutrinos in a previously unknown way—one that doesn’t match the standard models.”
So expect more “Dark XXX” fiddling…
The simple fact is “Dark XXX” fiddles have three things in common
1, They predict nothing.
2, They can be twiddled to match anything.
3, They are therefore “unfalsifiable”.
Which putts them in “chocolate fire guard” territory.
Some are getting tired of what they see “as this nonsence” and think that the area of science has dropped into a rut along the lines of “Shut up and calculate” and the cognitive issue of “run with the herd”.
I’ve long been suspicious of “Dark XXX” because it’s notions don’t match much that we would expect from other laws of nature. So if a simple soul like me can think “it’s silly” then…
Well, someone has applied an interpretation of evolution theory to the formation of the universe that has no “Dark XXX” and does not need it, so it peaked my curiosity. Importantly it is also “predictive” and importantly relatively simple and elegant which are oft basic indicators of “fitness” that indicates they are high probability of being based on fact.
If you have a spare half hour, have read of it and make your own choice,
https://theeggandtherock.com/p/the-blowtorch-theory-a-new-model
lurker • May 30, 2025 2:18 PM
@Bruce
re Squid Galaxy NGC 1068
Not to be mistaken for the Squid Nebula Ou4, which actually looks somewhat more like a squid.
lurker • May 30, 2025 3:16 PM
@ResearcherZero, Clive, ALL
re: Greynoise – ayysshush
It’s a pity the final sentence in the report, also so very, very relevant, is tucked away at the bottom of a rather technical report:
“Alternatively, if you’re not a reverse engineer capable of checking this for yourself, get your ASUS router off the internet.”
Roger H. • May 30, 2025 11:10 PM
A special Cup Noodle squid fork: http://blog.livedoor.jp/divajoanne/archives/1082779390.html
Clive Robinson • May 31, 2025 2:33 AM
Do AI Salesmen lie about AI?
It’s a question that has recently been raised over the so called,
“white-collar bloodbath”
Prediction of Anthropic’s CEO billionaire Dario Amodei reported in,
As,
“[The AI]technology he and other companies are building could wipe out half of all entry-level office jobs … sometime soon. Maybe in the next couple of years.”
Is it true, well it depends so “Yes and No”.
The basic germ/kernel of the statement is true, entry level office jobs and importantly senior management jobs will either nolonger be needed, or done better by some form of “automation” and is historically expected as “force multiplier technology” becomes viable.
But what does that actually mean?
Well not mass unemployment, job types especial niche skilled jobs get replaced all the time as the old joke about being,
‘An Unemployed “Sagger maker’s bottom knocker”‘
indicates[1]. The specific job dies out, but that does not mean “mass unemployment” as new jobs appear to replace them. It’s one of the falsities behind most,
“Utopian or Distopian predictions”
Because jobs are part of life and life is all about “Evolution in action”.
Back in the early 1970’s what we call “office productivity” was at it’s greatest and there were people saying that entire job types would be lost due to “Computer automation”.
The reality is skilled jobs did disappear, stenographers and typing pools are mostly things of the past. But the actual “real tasks” fell on other people as a new part of their job of having a computer on their desk.
Two things were evident from this,
1, There was no mass unemployment, in fact a skills shortage was created by new job types.
2, The net result was labour costs rose and “real” productivity actually fell, and it’s still not been regained to the same level after half a century.
The fact is the “new job types” don’t actually add to real productivity…
There is a reason I say “real productivity” not just “productivity” because it’s mostly a myth when it comes to “office work”. Because office work has significantly unneeded “repetitive drudge work” that mostly serves no “real purpose”.
One such can be seen by data collection and collation that gets put in what was once called,
“A data warehouse database.”
And now is a big chunk of the “cloud” “Storage as a Service” function that hemorrhages money out of a business for next to no return and increased risk and liability[2].
The purpose of collecting such data originally was “Business Engineering” that arose from the idea that many came to call “Time and Motion”. That in reality was fairly pointless because staff knowing they were being timed on “piece work” used to make the work look way more complicated than it actually was and thus “stretch the job out” to keep the “piece rate” they were paid high. Something we now call “gaming the system”.
However “gaming the system” is a classic case of “short term thinking”. In that management see high labour costs so replace people with machines and those people who gamed the system loose out. Or worse management “off shore” the work abroad.
But the people who most “game the system” are those in senior management. All rational independent analysis shows they have no basis to make the claims they make or the salaries and bonuses they claim. And don’t think they don’t know this, it’s one of the reasons “Business Consultancy” is still significantly on the rise.
Business consultants basically provide two things,
1, The answer wanted not that is needed.
2, Arms length protection and scapegoats.
And they likewise know it so they in turn look for the same two things.
In short it’s a “Hot potato game” and,
“It’s the last man left holding, who either gets nothing or burned.”
Which gives rise to two questions,
1, How does it work?
2, How will AI effect it?
To answer the first question, it actually mostly does not work “successes get claimed”, “failures get hidden” or where disclosure is unavoidable “blame gets externalised”. If you think about it we rarely get to hear of actual real “verifiable” successes, because the reality is it’s a lot less than “one in ten” senior management actions that actually show “real shareholder value” over/above industry average performance.
What happens is a senior manager likes to “imprint themselves” on the organisation. So they come up with “notions” –often from mantra– that the decide they will implement. Thus they have to convince others that their –usually– “hair-brained idea” will bring in significant ROI.
How do they do this, well you might have heard the quote,
“There are three kinds of lies: lies, damned lies, and statistics”
That gave rise to a book using it in part as a title,
https://www.hbs.edu/faculty/Pages/item.aspx?num=13928
The book indicates how you can come to almost any sales, mantra, or politically motivated conclusion you want by selecting the way you collect and collate data and apply statistical methods.
Yup pick your conclusion and massage the view-point of the data to match, thus show the “hair-brained idea” is “Better than a sure thing”…
I’ve mentioned in the past, how as a middle manager you can always be a success by “jumping ship early”.
So to answer the second question,
Current AI LLM and ML systems are also known as “stochastic parrots” both of which hide that the systems are actually little more than just glorified,
“Combined databases and statistics engines”
Which you can use to do exactly the same false presentation of data. Only they can “hide it better” due to the so far “black-box nature” of the encoded database.
That is actually what is being sold…
And just as with the “Personal Computer on every desktop revolution” did,
“Special skills” of certain job types that are effectively “drudge-work” will become obsolete due to the AI acting as a “force multiplier”[1]. New skills will be needed because of that will create new job types and real productivity will actually fall a couple of percent…
And so either or both the economy will have to grow in real terms or inflation happen to offset it.
[1] There will be people that will say “what the heck is?” In brief it’s an entry level job given to boys as a starting position as a form of apprentice role to take “drudge-work” off of the hands of more skilled workers thus making the more skilled workers more productive. However “force multiplier” machines and changes in production processes have more or less completely removed the need for the skills more than a life time ago. You can see more at,
https://www.thepotteries.org/bottle_kiln/saggar.htm
[2] In fact a lot of the data collected and collated is as @Bruce has noted in the past “Toxic”. And has high risk of liability to the “data owners” due to mistakes by the “Storage Service Suppliers”. With changes in legislation and regulation becoming an actual “existential risk” to some types of businesses using and dealing/trading such “Toxic Assets”.
Clive Robinson • May 31, 2025 6:09 PM
RFC4086 is twenty years old.
Titled : Randomness Requirements for Security
https://datatracker.ietf.org/doc/html/rfc4086
I was reminded of it today and people should realise that now “14 generations later” in many ways it is “very out of date” and really should be “brushed up” or replaced.
The intro starts,
“This Best Current Practice document describes techniques for producing random quantities that will be resistant to attack. It recommends that future systems include hardware random number generation or provide access to existing hardware that can be used for this purpose.”
Sounds good, but…
Most hardware these days is in effect either “unavailable” or “unreliable” at best.
It gives several recommendations like hard drive timing perturbation… Only Solid State Disks are effectively useless for this.
It also suggests “Ring Oscillators” these have always been a very bad idea as I’ve indicated long in the past and several times since on this blog.
Whilst they look random “close in” the reality is the output is a 1bit sampled sine wave at the difference frequency which can be observed by using an old fashioned analogue oscilloscope and turning the time base down. Or you can pull out a piece of graph paper and plot it out yourself.
There is so much wrong with it that it really needs to be bined and rewritten and kept updated every five years at most.
Because if you follow quite a bit of the advice in it you will not be secure…
Clive Robinson • June 1, 2025 3:08 AM
@ ALL,
In my above titled,
“Do AI Salesmen lie about AI?”
I put a link in to the US article I then quoted, but it did not make it for some reason.
However since then I’ve found other articles such as this “The Register” one,
https://forums.theregister.com/forum/all/2025/05/29/anthropic_ceo_ai_job_threat/
The main article says similar to the US article,
But It has comments…
That are a hoot to read and actually better than the comparatively staid articles 😉
@ Bruce, ALL,
I noted awhile back that two technical terms of art in a domain were,
1, Soft Bullshit
2, Hard Bullshit
The first meaning “unintentional lies” what in the case of LLM/ML systems also gets called “hallucinations”. The second meaning “intentional lies” which can be done with LLM/ML systems buy “false collation” of input corpus data and deliberate “prompt engineering” to get a desired answer, or by using one LLM to generate input corpus data for another LLM. Which is a form of “resonator effect” we call “echo chambers” when humans do it online.
Importantly the second also can be used as a shield by the corrupt to have a new form of “The computer says NO” that allows deliberate bias and mantra that the likes of Corrupt politicians and CEOs to as I’ve noted before “Arms Length” themselves away from their real intent.
Well there is another technical term of art some may have heard already but I suspect will get more prominent with LLM/ML systems failing the way they do.
On an ascending scale of uselessness we already have,
1, Drudge Work / Jobs.
2, Make Work / Jobs
And now
3, Bullshit Work / Jobs.
From which two questions arise,
1, Are academic professionals increasingly “Potty Mouths” ?
2, If “custom and practice” makes “Bullshit” normalised thus not a word to bring across invective, what word will replace it when vituperation is felt necessary?
But another issue with all such work that is on this expanding scale is it’s repetitive nature… Which moves it from the conscious mind and fatiguing effort, to the semi autonomous parts of the brain where “muscle memory” resides. This allows the conscious mind to be “freed-up” and other mental activities to happen. This has security/safety issues. The obvious ones being “click through” with OS and Browser warnings and “driving whilst distracted” that also applies to the operation of all potentially “body mangling” machines/systems.
Winter • June 1, 2025 5:00 AM
@Clive
“Do AI Salesmen lie about AI?”
Yes, and no.
Personally, I expect a rerun of the first digitalization of business.
When we were young (both of us), an office job most of the time involved sitting behind a desk typing out paper forms into other paper forms. More generally, it was 90% shuffling and stamping papers. Oh, and talking about paperwork over the phone to other office workers
Then all these papers were entered into computers and office work became shuffling digital documents and working spreadsheets. All this became more and more automatic. More and more could be done by the same number of people.
Now we are going the next step. A lot of simple decisions can be made by large statistical models like LLMs. Simple summaries and reports can be drafted by LLMs and then finished by a human. People can be more productive with AI, like they could become more productive with spreadsheets and word processors.
AI will be the Office Suit of our times.
There are a lot of ludicrous ideas in the sales talks, like AIs talking to each other over the phone. We have perfectly good network protocols for that. Or human-less factories (seeing is believing).
What I do expect is a rush to grab all the profits of the productivity increase due to AI by the powerful, like MS was able to siphon off a large chunk of the profits of the office automation rush in the 1980s/1990s.
You see it happening before your eyes.
But as there is already a growing labour shortage, there will be enough jobs to be filled. The fight will be about whether they will actually pay enough to live a decent life.
The prospects for the US are dire. They are going the way if their beloved and admired Russia. That is, the US is becoming a kleptocracy.
lurker • June 1, 2025 9:07 PM
@Winter
“Or human-less factories (seeing is believing).”
see also:
‘https://www.youtube.com/watch?v=_bVaMUds5gw @time 30’16” & 33’18”
That was seven years ago. I can’t locate another clip I saw back at that time of the BYD factory manager walking down a silent assembly line on a holiday, saying “It’s a pity the law makes me close the factory on this holiday. I would like to switch off the lights and go home, leaving the machines to carry on working.” And these are the jobs POTUS wants to “bring back home.”
ResearcherZero • June 1, 2025 9:17 PM
The false flag and disinformation operations of Unit 29155 hacking team.
‘https://theins.ru/en/inv/281731
ResearcherZero • June 1, 2025 10:21 PM
@lurker
It’s hilarious because 99% of the people who own those routers are not reverse engineers, would never read that blog and almost every household one enters has a least one device with a serious vulnerability connected to the home network. The admin password is on the pinup board and the WIFI password is posted on the fridge in case anyone forgets.
–
@Clive Robinson
The leaked blueprints of Russian nuclear missile facilities have been on the internet for the last year. At least some 2 million procurement documents detailing the supply chains.
The company providing the IT security training for the security agencies has poor security.
Speaking of which …A lack of guidance could lead to dangerous developments. If they make sensitive databases available to LLM and ML, security and safety measures might be wise.
‘https://www.wired.com/story/donald-trump-ai-safety-regulation/
Despite Trump leaving security to the states, he is also removing that ability…
“No state or political subdivision thereof may enforce any law or regulation regulating artificial intelligence models, artificial intelligence systems, or automated decision systems during the 10-year period beginning on the date of the enactment of this Act.”
Clive Robinson • June 2, 2025 2:58 AM
@ Winter, All,
(And in reverse order again)
You note,
“But as there is already a growing labour shortage, there will be enough jobs to be filled. The fight will be about whether they will actually pay enough to live a decent life.”
This is one of the things that makes me scratch my head about US –and increasingly other nation– Corps and their shareholders, and why I call it “short term thinking”.
At the end of the day their is a direct correlation between,
1, Individuals income
2, Their ability to purchase goods
3, A Corps ability to sell goods
4, The state of the National economy
It functions as a causal chain from the “feed stock” producer down to the individual consumer. Each of those chains forms a tree like structure with a time delay for each step down the chain.
So the higher an individual’s disposable income not locked into savings[1] or investments[2] the more purchasing power they have thus the more vibrant the economy.
It takes no real stretch of the imagination to see this makes every employee in a Corp the Corps eventual customers, even if they never directly buy the Corps products (that tractor, is purchased by a farmer who uses it to produce food, the Corps employees buy).
So cut the employees or cut their renumeration and eventually the Corp ends up cutting it’s own throat.
Worse as I’ve said for years here is “out-sourcing” and “off-shoring”. Both take out national economic churn thus very adversely effecting the national economy but also strengthening another nations economy.
I won’t go into “tarrifs” and “Sovereign Investment funds” because most people either get very political or their eyes glaze over… Let’s just say neither do what their promotors claim, and have significant downsides.
Which brings us onto your earlier observation of,
“There are a lot of ludicrous ideas in the sales talks, like AIs talking to each other over the phone.”
Yes and no, it depends on many things, but in the US something like 80% by area has no real communications other than the old dial up / POTS system that is optimised for voice at best. And if the major US Telco Corps have their way it will get worse even though the US Gov pays them “$billions” to supposedly change that. Basically the money gets spent on other things and pays inflated shares dividends, C corridor bonuses, and lobbying for the next “handout”. You and I might be nice and call it “grift” but the reality is it’s fraud by collusion. Or as it’s also known in the US “pork politics”.
It’s why even people who hate “Hellon Rusk” and his “throw the toys” behaviours are buying Starlink terminals and paying monthly fees that would shock many Europeans.
And it’s not just the US most of Australia by area has not even POTS phones but uses HF and VHF Radios. The same is true for South America and Africa and large parts of the area between the Eastern edge of Europe and across to the Pacific where one prevalent attitude is theft of infrastructure is a legitimate way to feed your family. In the US and some parts of Western Europe it’s known as “scrapping” and is said in Europe to be “Serious Organised Crime” and as you’ve probably guessed “attributed to Eastern European Crime Gangs” by certain MSM rags and their ilk.
The human voice is at the end of the day “low bandwidth” but comparatively “highly fault tolerant” and will work on just about any low level interconnect system with next to no technology required… If you have to have AI Voice for “Real Life People”(RLPs) then using it for LLM to LLM communication should in effect “come for free” (but we both know it will have a massive “rent seeking” monthly cost, such is the nature of the “Corporate Beast”).
[1] The economy also needs “savings”, as these should go into “future purchasing” by the individual. But in the mean time the “money” is loaned to Corps etc to build production thus eventually the economy.
[1] It’s said the economy also needs investment like shares. I see shares as a dubious thing, because they are a variation on “forever debt” much like “interest only repayment loans”, the theory being inflation devalues the capital debt but that is untrue of shares. It’s not much talked about but one of Europe’s largest retail chains has shares held “only in the family trust” and they raise capital when they need it by long term low interest loans mostly by the equivalent of mortgages on property they already own. Thus all we really know about them is what they chose to reveal which is very little.
Clive Robinson • June 2, 2025 3:18 AM
@ ResearcherZero, ALL,
With regards,
“No state or political subdivision thereof may enforce any law or regulation regulating artificial intelligence models, artificial intelligence systems, or automated decision systems during the 10-year period beginning on the date of the enactment of this Act.”
That as written, is “utter madness” because that is “blanket immunity”. So if I create a system that is actually designed to kill as many US citizens as possible and put “alleged AI” in as the targeting system… then no legislative or regulatory body can take action to stop me or my “Kill-Bot” ={
I’m just wondering when “Trumpian” will replace “Faustian” in common parlance and dictionaries world wide.
Bob Paddock • June 2, 2025 9:02 AM
@ Clive Robinson
“… So expect more “Dark XXX” fiddling… …”
In a Plasma/Electric Universe model there is no need for Dark Anything.
The Thunderbolts Project has been covering this science for a long time:
‘https://www.thunderbolts.info/wp/
Clive Robinson • June 2, 2025 2:26 PM
@ Bob Paddock,
I find I don’t trust the astronomical “Standard Model” as it’s suffering a death by a thousand cuts as our knowledge improves from what we observe that conflicts with it.
Further much of the corrections are as I’ve noted, are not predictive, and the consequent fiddles make testing by the process of “falsification” effectively not possible in effect they are turning things up to eleven and beyond… Which is a similar failing to some of Einstein’s models.
But the Electric Universe model is to put it politely far from complete. However it’s view and similar models are predictive. And so far a test has not been found to show it is false.
But I have a more fundamental issue which is the evolutionary start of,
“Where did we come from?”
At the most fundamental level of energy and matter.
For instance though not much talked about we now have knowledge that says that the Universe is not bounded and energy/matter is not constant. Thus one of the most fundemental axioms of the laws of nature is being challenged.
I’ve suspected things are wrong for some time as you will find on looking back through this blog I say,
“The laws of nature/physics as we currently understand them.”
Another is the issue of “Cause and effect” as has been indicated “rabbits do not just appear in hats” but “nor is it magic”
For various reasons Fred Hoyle thought the “big bang” to be incorrect and I only found out after reasoning similar myself.
What I can say is that “times arrow” is either wrong, or we can not look back beyond a certain point in time.
I’ve likened the “Big Bang” to being the equivalent in argument of a “pin hole camera” where the smaller the hole the sharper the image.
On the assumption the Big Bang is in effect correct, how do we know it’s not the only one? It was once called the string of pearls notion of succession after succession of “Big Bangs” happen… But the question then arises of “The big crunch” where everything collapses down to go through the next pin hole. Not exactly a satisfactory answer, because a fundamental tenant of “cause and effect” is there should be a point ot origin even if it does collapse down to a ring of behaviour.
So the,
“Where, When, and Why?”
Questions remain not just unanswered but avoided, usually to the notion of an “outside our universe / super entity” which again avoids the questions but adds a fourth of
“What, purpose?”
Firstly I don’t “believe in deities / super entities” as it smacks of “excuses to children” at best and the charlatan game of “divine right” to hide the mental aberration of “might is right” and all that goes with it at worst.
I just think humans should actually believe in other humans and what we collectively can achieve together rather than trying to destroy each other and all that is around us. As Carl Sagen and others pointed out, for all it’s vastness we are the only place in the universe we know has life on it, thus perhaps we should take better care of it and what’s on it.
Secondly, another result / argument of which has been the “many universes model / theory” which also gives us the “swiss cheese model / theory” and so on. The many universes notions whilst intriguing in effect argue for infinite “somethings from nothing” which if you think about it for long enough is kind of perverse.
So I’m doubtful about all of them, though I do like esthetically the “Universe as a computer” argument because it does give an answer toward “purpose” we can understand.
But the human quest to “know” which supprisingly to many drives us beyond what we’ve observed in many creatures –or failed to see/recognise– and is mostly the underlying cause of “society” remains unsated thus we remain driven forward.
Which is why one of the observation on the site you link to remains true and in effect references the four W Questions of “Where, When, Why, What” underlying “origin and place within”,
“Independent researcher Stuart Talbott defers to Wall Thornhill’s position that the EU Model does not need to complete with the Standard Model because the humbling reality is that we do not yet have sufficient understanding of the origin of the Universe or our place in it.”
So the quest to chart the unknown path continues.
lurker • June 2, 2025 5:27 PM
@Clive Robinson
“And so far a test has not been found to show it is false”
And there all our theories fall: assumptions based on our inability to prove a negative.
not important • June 2, 2025 5:31 PM
Are the Systems We Have Created Now Controlling Us?
https://www.psychologytoday.com/us/blog/word-less/202503/are-the-systems-we-have-created-
now-controlling-us
=We face a future where we’re totally dependent on technological systems, treated as data, not humans.
When you step off, a street facial recognition system scans your face, cross-referencing your digital ID with crime records, passports, and banking.
Today, systems thinking permeates every aspect of life. Our personal lives are mediated and monetized through digital systems, algorithms, and data. We’ve become data-generating entities in the “information economy” in the “fourth industrial revolution.”
Today, Big Tech has taken the belief that the human being is a mechanistic system to its
conclusion, promoting the idea that merging humans with machines is not only possible but
inevitable. This belief drives hundreds of billions of dollars in investment in the
technologies of brain-computer interfaces (BCI), cybernetics, and AI. The goal is to achieve sentient machines by merging humans with computer systems, even if that means lowering the legal restrictions on experimenting on human test subjects or lowering the standards of human intelligence so that AI can claim to have reached human-level intelligence (AGI).
To resist, we must first ask ourselves: Are humans more than the sum of 11 biological and
five mental systems? If so, we might begin by examining our dependence on these systems. Can we live without them? Might we have the courage to turn off our health trackers, silence our AI assistants, and feel our own living pulse?=
The Question is who control those systems behind the scene directly or by providing biased training data?
not important • June 2, 2025 6:18 PM
Infrared contact lens enables humans to see in dark
https://www.dw.com/en/infrared-contact-lens-enables-humans-to-see-in-dark/a-72749143
=Chinese researchers have developed an infrared contact lens that makes night vision
possible. Nanoparticles make the previously invisible light range visible to the human eye.
The human eye can only perceive a small section of this spectrum, approximately the range between 400 and 700 nanometers.
Because of that, we humans are unable to see the infrared range, with its longer wavelengths of 750 nanometers to one millimeter.
Researchers at the University of Science and Technology in Hefei, eastern China, have now
developed a contact lens that converts infrared light into visible light, enabling humans to see in the dark.
Yuqian Ma and his team have combined conventional soft contact lenses with 45 nanometer particles consisting of gold, sodium gadolinium fluoride, ytterbium and erbium ions.
In tests, humans were able to recognize patterns, letters and flashing infrared signals in the dark. And the infrared lenses work even better with closed eyes, because the infrared light can easily penetrate the eyelids and image generation is not disturbed by normal visible light.
Several animal species are able to perceive infrared light, which is extremely helpful when hunting in the dark. They do not see infrared light as “light” in the sense of human vision.
Instead they perceive the heat radiation emitted by objects.
According to the developers, the lenses could be used in surgical procedures, in the
field of encryption or cryptography, or for counterfeit protection.
This is because infrared light is what makes invisible features or inks visible on
documents, for example.
The lenses could also be used to rescue people in poor visibility conditions because they make heat-emitting objects visible. However, many critics doubt this, as night vision devices are much easier to use, and are also significantly more powerful.=
Dancing on thin ice • June 3, 2025 1:55 AM
Isreal secrets shared to Russia in 2017 and some of their other intel mentioned in the leaked Signal chat led to speculation of when other countries may be reluctant to share intel.
Plans for taking out of 30% of Russia’s aircraft were kept under wraps amid the numerous incidents of flauting basic security protocols by the current United States administration.
Granted, plans started a year and a half ago, but that would have been while election campaigning was underway where intel would be shared with both candidates.
Bob Paddock • June 3, 2025 10:31 AM
@Clive Robinson
“I find I don’t trust the astronomical “Standard Model” as it’s suffering a death by a thousand cuts as our knowledge improves from what we observe that conflicts with it.”
Yet The Establishment will not let it go no matter what data is presented.
“Electric Universe model is to put it politely far from complete.”
True. That is why it is a model. It does solve many of the problems with the Gravitation centrist Standard Model.
“… Universe is not bounded and energy/matter is not constant. …”
Are you familiar with Rupert Sheldrake’s Morphic Resonance and Morphic Fields?
‘https://www.sheldrake.org/research/morphic-resonance/introduction
“For various reasons Fred Hoyle thought the “big bang” to be incorrect”
Nothing about the Big Bang has every made logical sense. If the Big Bang was unknown to us right now and a five year old proposed it, would it be accepted? Probably not.
Questions like what did it Bang into? Infinite void? What powered it? Why at that moment? How do we know there was only one (which you mention)? As well as many others go unanswered.
“What I can say is that “times arrow” is either wrong, or we can not look back beyond a certain point in time.”
Retro-causality has been proven in many experiments. I’ve brought them up here in the past, not that long ago. Volumes of that research data, as well as the equipment, is now in the UK if you want to study it.
Decision Augmentation theory purports that every decision we make is influenced by our future.
Optical Phase Conjugation is also interesting due to its use of apparent negative time.
Concetto R. Giuliano, “Applications of optical phase conjugation,” Physics Today, vol. 34, no. 4, pp. 27-35, Apr. 1981.
Abstract: Light waves that are, in effect, time-reversed images of their original can serve to restore severely aberrated waves to their original state.
“I do like esthetically the “Universe as a computer” argument because it does give an answer toward “purpose” we can understand.”
So we are living in a Simulation?
Are you familiar with Tom Campbell’s “My Big TOE (Theory of Everything)”? He describes your Universe Computer.
“So the quest to chart the unknown path continues.”
Yes. Saying “I do not know” is the first step in learning anything.
“What, purpose?”
I do not know…
Clive Robinson • June 3, 2025 8:14 PM
@ ALL,
Is Microsoft doomed in very near future?
There was little said about Microsoft’s criminal “sanctions busting” behaviour in Russia in either UK or US Main Stream Media or for that matter trade journals.
Then Reuters came up with the news that Microsoft Russia had become bankrupt,
https://www.reuters.com/technology/microsoft-unit-russia-file-bankruptcy-database-shows-2025-05-30/
Getting further “reliable” details has proved difficult.
In the main because of,
1, Win 11 being a disaster and having no chance of getting out the door in a “fit for purpose” state.
2, Microsoft’s over investment in AI and not being able to “push it onto customers” at even below cost is causing investor twitters.
3, Microsoft rapidly pulling out of, or significantly back peddling on server farm space acquisitions (allegedly that was ear-marked for AI racks). Causing speculation about Microsoft AI efforts/investments tanking.
4, Microsoft having yet another major layoff of staff, appears this year to be a monthly event, again causing investor twitters.
5, Microsoft Bing still being a disaster in free fall, as well as a bad advert for Microsoft AI.
6, More rumours that Microsoft will leave by far the majority of Win 10 users in the dust as the users existing hardware won’t support Win 11 in multiple different ways.
7, And the rumour mill is saying that the Cloud based “office” products are leaking confidential information that Russian entities have been hovering up…
8, Official figures showing Linux –via Android– now has a greater number of users than all the currently supported MS OS user installs…
9, More and more “Tech Journos” publishing articles saying migrating to Linux is a better option for nearly all users than Win 11 (even for those users whos hardware will support Win 11).
And several more you could pick from to make a “Top 10 Doom list” with.
Even supposedly “good news” for Microsoft is tinged with the wiff of desperation and failure…
Oh and finally some Tech Developers waking up to the fact that “collective bargaining” entities (unions) are not something that should be “poo pooed” or derided any more.
Time to go and buy a sack of pop-corn and a spare microwave as I really can not guess at how many bowl fulls will be consumed by those watching from the side lines…
Winter • June 3, 2025 10:40 PM
@Bob
Questions like what did it Bang into? Infinite void?
Space is created together with the energy/matter. Without energy/matter there is no space.[1] So there is nothing to bang into. There is no “outside” of the universe (as the name implies).
What powered it?
Best guess, there was something [2] in a non-lowest energy state that changed to a lower energy state, releasing the energy. However, physics, and logic, cannot explain the creation of something out of nothing. So it cannot explain why there was something to begin with.[3]
Why at that moment?
Time was created together with space (space and time are not separate things in relativity). So there is no time before the big bang. The precise moment of creation is undefined if there is no before.
How do we know there was only one (which you mention)?
We don’t. However, there are as yet no known ways different “universes” could affect each other in ways that would enable one to detect their existence. So it might even be impossible to know whether there have been more bangs.
“I do like esthetically the “Universe as a computer” argument because it does give an answer toward “purpose” we can understand.”
Everyone has a right to make sense of reality in a way they see fit.
But the reason the universe can compute is that if a physical computer exists, all of physics must necessarily compute.
The second law of thermodynamics tells us that a non-zero temperature implies there is information that is transforming, evolving, over time. As these transformations are all unitary [4], they are equal to computations.
All this tells us nothing about whether there is something or someone who reads out the results of the “computations”. Or that, like Spinoza wrote, the universe itself is God and we are all part of it.
[1] Actually, the best theoretical physics can do yet is that space, distance, is determined by the quantum entanglement between “points”. That is, quantum entanglements are what we call space.
[2] A field, quantum field theory, but that is just a label and explains nothing.
[3] Actually, physics students are told at the start of their studies that physics is about calculating what happens, not understanding it. “Understanding” the universe are subjects of philosophy and metaphysics. In the symbolic world, mathematics is a better choice for a student who wants to understand.
[4] All of quantum mechanics is unitary (look it up). Therefore, all of physics evolves unitary. Thus the universe itself evolves unitary.
Clive Robinson • June 4, 2025 4:28 AM
@ Bruce, ALL,
Tariffs critically effect security
Is pretty much a given when you consider all “production chains” involve “cross border” trade.
In effect like the saying about man,
“No nation is an island”
In our modern world where resources are far from uniformly available, and likewise to be efficient production generally has to be “mass production”.
Thus local to a point which implies there has to be “supply chains” between the points of production.
Further supply chains consist of many points
between geographically dispersed “Holes in the ground and a consumer”
We know the only way to reduce the “holes in the ground” issue that cause so much environmental damage is the dreaded “R Word” of “recycling” (something nature does all the time, but Capitalists and apparently Consumers shun preferring to create other holes in the ground as a solution…).
Tariffs clearly interrupt supply chains as at best they add an “inefficiency”… So at the very least cause them to be changed or cease to exist. In the latter case this can mean production has to cease unless there is an alternative hole in the ground or sufficient recycling of materials to supply as “feed-stock”.
Thus individual “goods” that are not isomorphic with another “good” will cease to exist when the supply chain is in effect broken in some way.
A current example of this is,
The title is a little misleading because it makes people think it’s a “supply and demand” issue of “no demand”. However the reality is it’s “priced out of the market” due to the increasing cost of supply chains caused by tariffs or the fear of tariffs.
The economic counter argument is that “goods and production” are flexible thus one good can be replaced with another, likewise production.
Whilst that might be true in the long term it’s certainly not true in the short term. Because centers of production can take decades to establish. But also in the short term of no availability an entire market segment will probably cease to exist.
And that is a significant problem because new production centers in other geographic regions will not be built if there will be no market to sell products into that region.
There is another aspect I won’t break down, but one side effect of markets ceasing is another dreaded “R Word” and that’s “Recession”.
Something arguably the US is starting to increasingly see in just this year alone.
The net effect of a non global recession is “skill loss” due either to skilled individuals migrating to where there is work, or being unemployed in that skill causing the residual skill level to become out dated and fairly quickly uncompetitive or lost.
We hear of the “10,000 hours to learn a skill” what is not talked about is the “1,000 hours for a skill to be lost”. Yes it can take only six months for the loss of a skill or for it to become obsolete.
People who have had serious injury such as a broken leg will tell you that they “are never the same”. It’s known that as little as a broken toe can kill a dancers career, similar with athletes. Likwise with skilled people “the edge comes off” very quickly and may never be regained.
Thus we come onto “security” specifically for this blog “ICTsec” on the face of it as much of it is “mental not physical” skills people might think that “supply chains” are not relevant, but they are…
Because all of the mental skills are about physical constructs / objects. If you can not obtain the products or tools to make them, then you fall behind and your mental skills at best tarnish or move to alternative skills. Either way the game of catchup will be not cost effective for individuals or organisations. And tariffs will do the same for entire nations as well.
And post WWII industrial history will show you pretty much what will happen. Employment will migrate to where it costs least at any given skill level. Once gone it becomes subject to an inverse “ratchet effect” so is unlikely to come back unless some other factor like “innovation” causes a divergent path that in effect requires “new skills”.
You will find AI being talked up as that “innovation” that will make people so productive their costs will be lower than anywhere else…
It’s a bubble that needs bursting for two basic reasons,
1, Current AI LLM is totally inadequate for all but a tiny handful of skills replacements.
2, Any innovation these days is not local but global so confers no cost differential.
The few areas that LLMs can give an advantage are very niche at best, and usually need a whole production supply chain…
Catch-22 no production already in place means that the advantage goes to where there is production capable of benefiting from it. This implies “cutting edge production” that can take decades to acquire, by which time the innovation will nolonger offer any advantage.
Clive Robinson • June 4, 2025 7:05 AM
@ Winter, Bob Paddock,
With regards,
“But the reason the universe can compute is that if a physical computer exists, all of physics must necessarily compute.”
Hmm…
Computation as we know it currently manipulates information that is modulated / impressed on energy / matter. So is subject to what we see as the laws of nature.
However there are three things you can do with information in our physical realm,
1, Store it,
2, Communicate it,
3, Process it.
Each is dependent on the ability to do what comes before it in the list.
However storing information does not entail either communication or computation. Nor does communication require computation.
Whilst I can compute with an abacus the individual beads remain what they are and don’t store, communicate or process information.
Winter • June 4, 2025 9:57 AM
@Clive
However storing information does not entail either communication or computation.
A computation is taking a system who’s evolution models the system of interest, say, proposition logic. Then set up the system in the desired starting state and let it evolve. After the evolution is complete, the end state is recorded and translated back to the equivalent state of the system of interest.
“Information” is the equivalence between the states of the model system and the system of interest.
Winter • June 4, 2025 10:29 AM
@Clive
However storing information does not entail either communication or computation.
Continued…
Turing’s Universal Computer (UTC) showed that any digital “computer” can be emulated by any other UTM. As UTMs can be implemented by almost any physical system if enough complexity to implement a nand gate, we know that any physical system of any complexity can “compute”.
The same arguments have been raised for quantum systems and quantum computers (eg, The computational capacity of the universe, by Seth Lloyd).
Computational complexity is already used to model the evolution of black holes/wormholes (ER=EPR on the relation between entanglement and wormholes).
To go back to my initial claim, a computer can compute because any physical system can compute. However, what raw physical systems process is not “information” but “entropy”. And entropy exists independent of interpretation.
lurker • June 4, 2025 1:41 PM
@Clive Robinson. Winter, Bob
“Whilst I can compute with an abacus the individual beads remain what they are and don’t store, communicate or process information.”
So long as the individual beads remain where they are, their positions store information, like a register of !s and 0s.
One might argue that communication requires an observer to see the positions of the beads, but if a cable has an information device at one end, and nothing at the other, is there communication when the electrons (or photons) move in the cable?
The abacus requires an external force to move the beads as they process information, just as a computer requires a power plug.
Clive Robinson • June 4, 2025 9:31 PM
Copyright destroys Privacy
Many have assumed that their searches and enquires with AI tools were effectively ephemeral, thus lost to time thus having privacy.
In a court case involving OpenAI the court has ruled that all “chat logs” and other “user related” data be permanently retained.
After the copyright holders accused OpenAI of wantonly destroying evidence.
The case against OpenAI was brought by a group of news/media organisations on the “usual suspects list”
Firstly it needs to be pointed out that “news media organisations” are “reporting organizations” that is yhey do not create the news therefore they do not own the news. All they can lay claim to is how they “create a derived work”.
Their argument is in effect anyone using “news” they illicitly claim is theirs is committing some form of crime effectively theft and that Open AI is complicit in this theft.
We saw this same sort of nonsense from the Rupert “the bear faced lier” Murdoch organisations resulting in moronic legislation in Australia. At the time several people myself included concluded that it was the thin edge of the wedge and that it would embolden other legal ploys by the otherwise increasingly irrelevant news media organisations.
In essence they are trying to coral free speech and put a lien on it by claiming they own the news.
Thus if you were to tell me that the Whitehouse has a black cat wandering around that you had seen when visiting, and I ran a search on it… Murdoch and his evil empire and similar are claiming I could only have come to that query and the search tool provided answers through “their news”…
So kiss goodby to “free speech”.
But remember Murdoch and co are known to steal stories not just from other media/news outlets but in every which way they can like spying on people through surveillance techniques and bribery of people in places of trust (ie misfeasance).
This is intolerable behaviour in a free society and attacks the very foundations of both freedom and privacy to at the very least force “rent seeking” on “social intercourse”.
lurker • June 6, 2025 3:13 PM
Friendly Fire
Because of our low population density and irregular topography, cables are uneconomic in many areas. RF has become a popular distribution method for roadband internet. Until it gets jammed …
I recall an early shipborne aircraft detection radar of which I had intimate acquaintance. It was forbidden to power t on within 10 miles of our port because t took out the local TV channel.
Subscribe to comments on this entry
Sidebar photo of Bruce Schneier by Joe MacInnis.
ResearcherZero • May 29, 2025 10:36 PM
SSH backdoor can only be removed by factory reset and manual reconfiguration.
‘https://www.greynoise.io/blog/stealthy-backdoor-campaign-affecting-asus-routers
Vicious Trap
https://blog.sekoia.io/vicioustrap-infiltrate-control-lure-turning-edge-devices-into-honeypots-en-masse/
The SSH backdoor persists across firmware updates and uses no malware.
http://www.labs.greynoise.io/grimoire/2025-03-28-ayysshush/