This is what I was given a little over 5 hours ago when I was given (at https://www.bbc.com/news/articles/c5j9x9pr0240o) ‘Pentagon stops using Anthropic AI tools after blacklisting company, BBC told’ with the subtext “The US Department of Defence is no longer using Anthropic’s AI tools, an official has told the BBC, months after it designated the company a supply chain risk on national security grounds. The Pentagon “has ceased the use of Anthropic products”, the department official said in a statement on Monday.”
We can speculate all we want, the simplistic version that I reach for is that all AI is fake, it is an advance form of predicted analytics and as I was able to set the stage of ‘Maybe we bomb Iran to hell’ (or something of the sort states by President Trump) all whilst I needed nothing more that the advanced statistical tools seen below:
An advanced statistical took invented by Babylonians 4500 years ago. And it told be that the US will have depleted its stockpile before that happens and it needs 3-4 years to get their stockpile up to date (that is if they have the dineros to pay for that all) and there is more, because a war is won by grit (in the very least) and I reckon that the tools by Anthropic told them as much and what does this administration do when the outcome doesn’t match their reasoning? (Im leaving that in the open) But the setting of National security is a laugh to say the least. The end of the article by the BBC gives us “Hegseth designated the company a “national security supply-chain risk” – a law typically reserved for companies based in countries that pose a threat to the US.” So what is a ‘supply-chain risk’? Im just putting the question out there, because if these are tools specific for the defense department, so either it will not or does not give the answers they need. Optionally it makes Anthropic the juiciest tool in their tastebuds, but the distancing parties are at least as intelligent as I am and they see that beyond the marketing all AI is fake (at present) as such what danger is there to national security? At best it is (under optimal data settings) a decent predictive analytics tool, under debatable settings (like I heard from my coworkers in Chicago) that certain ‘people’ gave them the run around with excuses like: ‘we’ll fix this on the backend’ or ‘we are about to get there’ and I reckon they never got these points settled, it was too hard for them. As such their national security is riddled with the GIGO laws in place (Garbage in, Garbage out). Now this last part is presumption, because I was never in the know. But it is the most likely of settings and anyone deep enough with Anthropic (or OpenAI) could see that in a near instance and that is where the trouble starts. As I personally see it Dario Amodei has a backbone, a hardcover, not a paperback like Sam Altman (I merely accuse him of being narcissistic, which is not a crime) as such the defense department is likely making a few changes in the short term (or if they are merciful on their data departments they will cut these fake AI’s all together).
I would prefer if they do that AFTER they pay Oracle for their services, but I don’t work for Oracle, so I actually don’t care. But it would be nice for Oracle. I am a decent bloke in that regard. As such the hardcover mention comes from “But earlier this year, the Pentagon pressed it to remove its safety guardrails from Claude and give the military unfettered access to its AI products.
When Anthropic refused, citing concerns of mass surveillance and autonomous weapons” which is a likely culprit, but it in not the only reason. I reckon that any media worth its salt has been pouring over the data chaps in the Defense department for a while now and as they are not as heavily dined like the people at Langley they might spill a java bean or two during desert to one of those Washingtonian women in journalism, but that is just me postulating.
I reckon that there is some truth that this US administration would like to blame Claude for whatever comes next, even if there is enough data out there showing that the GIGO laws are to blame and the people in charge are ‘assuming’ that the data is flawless when it wasn’t even close in the late 90s and in the 20 years that followed the mess got greater. Because from two faulty assumptions a truth can never be distilled and that is the one direct consequence why I am so intent on verification and validation. All results derive from these two truths in action and in place and as far as I can tell none of these systems have this, as such the predictive analytics are prone to errors, the fake AI can never derive anything from validated data and if there is no verification the faults merely increase. So what happens? Mass surveillance goes wrong because the the subject was non-caucasian and that is before we get the setting that translation errors of a passport come into play, the simple validation of a passport, a human requirement and all that is up for grabs as we consider the BBC article where Pete Hegseth is looking at the President like he is admiring ‘his Masters Voice’ (older readers will get that reference).
And just like that we see another ‘failure’ (in a good way) hit these fake AI (or Super Intelligence) hit the world. So where are you in all this? I am merely asking, because we have a multi billion dollar device being outclassed by a $30 device not needing electricity. Did I oversimplify the problem here? Have a great day today.
