amanfromMars 1 Wed 26 Apr 09:08 [2304260908] …. outs Secret ProgramMING on https://forums.theregister.com/forum/1/2023/04/25/ot_ethos_critical_infrastructure/
Who Dares Win Wins ….. with Special Advanced IntelAIgent Research Services and Alien Sources.
Once up and running, “ETHOS will collectively uncover and share emerging threats for which there is no threat intelligence or no known attack pattern available, across private and public sector stakeholders,” the ETHOS Community said. It describes ETHOS further as an always-on, hotline-esque system that will “correlate info from many security vendors to identify anomalous behavior.”
The emergent and expanding and strengthening difficulty/problem for the likes a CISA/DoD/Dept. of Homeland Security/public Government body, and to deny such is a current colossal vulnerability and global hazard for proactive exploitation and employment in successfully comprehensively betatested applications by ethereal A.N.Others, is whenever the likes of an ETHOS discover emerging threats from the virtually and practically untouchable exercising novel creatively disruptive information and exclusive proprietary intellectual property share [aka top secrets] in deep and dark webbed private and pirate body sectors on which there is no threat intelligence nor any known attack pattern available for effective defence of the conventional and traditional and status quo private and public sector stakeholders, are a great deal more viable and attractive than anything being proposed and live tested/trailed and trialed experimentally elsewhere.
…………………………………..
amanfromMars 1 Wed 26 Apr 14:46 [2304261446] ……. asks on https://forums.theregister.com/forum/all/2023/04/26/kemba_walden_cybersecurity_space/
First things first ….. cui bono?
US National Cyber Director: Fending off cyber threats in space is ‘urgent,’ needs ‘high level attention’
I would ask the US National Cyber Director whether he is flogging a dead horse in trying to defend the indefensible and inequitable and generally unpleasant and universally despicable …. for such is always going to be an increasingly expensive and self-destructive lost cause?
A colossal folly, no less.
…………………………………………
amanfromMars 1 Wed 26 Apr 18:14 [2304261814] …. points out on https://forums.theregister.com/forum/1/2023/04/26/simon_willison_prompt_injection/
Some points to reconsider and re-evaluate, Thomas/Simon
but really, the underlying language model is far more capable: it’s just constrained by this so-called prompt engineering.
The prompt engineering is being used by SMARTR bots to better train following fans and AIgents with exposure to their responses able to be acted upon and subsequently realised.
It gets worse. With large language models, anyone with a keyboard is a potential bad actor.
Others would suggest, and be both willing and able to prove, that things can become much better whenever anyone with a keyboard is a potentially great than normal actor.
And is that not just copying what human developers have been doing since time immemorial began. It aint nothing new, is it, so just get it on with IT and quit fcuking around and wasting everyone’s time in what you are discovering are completely different spaces to what you have been used to commanding and controlling ie they don’t necessarily listen to your input/output.
……………………………….