Since some people seem to not be familiar with Joanna, she is a massively influential security researcher. IMO most famous for her "Blue pill" attacks on Vista and Xen: https://blackhat.com/presentations/bh-usa-06/BH-US-06-Rutkow..., https://invisiblethingslab.com/resources/bh08/part3.pdf. Her work demonstrated that hardware virtualization is not in fact the security panacea we wish it was, but that it too is vulnerable to attack just like any other layer of the stack. This revelation resulted in significant changes in the security world.
She is also famous for the Qubes OS project and coining the term "evil maid attack".
> "Blue pill" attacks on Vista and Xen [...] Her work demonstrated that hardware virtualization is not in fact the security panacea we wish it was, but that it too is vulnerable to attack just like any other layer of the stack.
Blue-pilling is a method for malware to hide from the OS by virtualizing it, not an attack on VMs. That's why it's called "blue pilling". I do agree though that VMs are not airtight and VM escapes have been demonstrated.
Yes, but when you blue-pill the hypervisor it becomes an attack on VMs. Her demonstrating the blue pill attack on Xen was really the "oh crap, these aren't as safe as we thought" moment.
Kudos to her pristine work in security, especially with QubesOS - more relevant as ever since LLMs are now finding and weaponizing bugs at lightspeed, QubesOS feels like the last trustworthy OS fortress left.
> What is Qubes OS?
Qubes OS is a free and open-source, security-oriented operating system for single-user desktop computing. Qubes OS leverages Xen-based virtualization to allow for the creation and management of isolated compartments called qubes.
These qubes, which are implemented as virtual machines (VMs), have specific:
Purposes: with a predefined set of one or many isolated applications, for personal or professional projects, to manage the network stack, the firewall, or to fulfill other user-defined purposes.
Natures: full-fledged or stripped-down virtual machines based on popular operating systems, such as Fedora, Debian, and Windows.
Levels of trust: from complete to non-existent. All windows are displayed in a unified desktop environment with unforgeable colored window borders so that different security levels are easily identifiable.
> This blog is intended to be a reportage of my struggles between Rationality and Humanism. Pragmatism and Beauty. Formalism and Intuition. Freedom and Love. Individualism, Independence and Privacy on the one side, vs Egalitarianism, Sharing and Community on the other.
What is he on about? Sounds like his future blog posts could just be rambling about anything.
Has she inspired any rationalist polycule harry potter fan fiction enthusiast cults that live on a decaying tugboat before murdering several people yet?
I mean, 'mentions Claude while asking it about AI consciousness at the end of long post about consciousness'**. Seems fair?
Apologies if I'm misreading 'incentive' and missing some jargon usage, ofc! Or if this just a lament that she's not a purist/gold star doomer? Cause I totally understand that.
Joanna, if your reading this, welcome back .. really miss your blog !
In my ideal future, Joanna would write about things like :
- how do we secure LLMs leading up to AGI, presumably they can bypass our best attempt at airgap firewalls
- is consciousness different from thinking from intelligence ?
- how do we ensure AI is distributed so all humans can benefit, and not used as a multiplier of extreme wealth inequality ?
- whats missing in current AI LLMs ?
I'll take a crack at the last one : imo, the current generation of LLMs is missing :
- a bias for truth / fact .. or a mode switch to make it bias truthiness
- reasoning by chains of formal language [ formal logic ]
- reasoning by chains of probabilistic inference [ bayesian logic ]
- reasoning by deep simulation [ stochastic modelling ]
- spatial reasoning [ 3D model of space, machinery, physics. 3D reconstruction. model of humans and animals with bones, muscles, mass ]
- mathematical modelling [ proposing formulae, checking fit ]
- psychological model of humans, of human populations, their needs / motivations / rewards and psychoses
Essentially AGI would require a proper merging of RL style [ NN learning from a stochastic simulation of future states ] and current ChatGPT style LLMs.
An implication of the above is that future AGI will need to run on fast branching CPU _and_ massively parallel GPU with a fast data path between them - ie. balanced compute.
However, the best part will be discovering what the real Joanna writes about !
She is also famous for the Qubes OS project and coining the term "evil maid attack".
Blue-pilling is a method for malware to hide from the OS by virtualizing it, not an attack on VMs. That's why it's called "blue pilling". I do agree though that VMs are not airtight and VM escapes have been demonstrated.
> What is Qubes OS?
Qubes OS is a free and open-source, security-oriented operating system for single-user desktop computing. Qubes OS leverages Xen-based virtualization to allow for the creation and management of isolated compartments called qubes.
These qubes, which are implemented as virtual machines (VMs), have specific:
source: https://doc.qubes-os.org/en/latest/introduction/intro.html> Today I no longer do computer security. I do other stuff, part of which running this blog :-)
What is he on about? Sounds like his future blog posts could just be rambling about anything.
She was a brilliant cybersecurity researcher.
Nowadays seems to have switched to poetry. Not the first one to do that, recently...
IMHO she is up there on par with Yudkowsky.
Has she inspired any rationalist polycule harry potter fan fiction enthusiast cults that live on a decaying tugboat before murdering several people yet?
https://www.google.com/search?client=firefox-b-d&q=wikipedia...
https://tracesofhumanity.org/freediving-embodiment-and-human...
Please, please, please let this not be the incentive!
Apologies if I'm misreading 'incentive' and missing some jargon usage, ofc! Or if this just a lament that she's not a purist/gold star doomer? Cause I totally understand that.
wild assumption. god forbid someone look for some humanity in this world.
In my ideal future, Joanna would write about things like :
I'll take a crack at the last one : imo, the current generation of LLMs is missing : Essentially AGI would require a proper merging of RL style [ NN learning from a stochastic simulation of future states ] and current ChatGPT style LLMs.An implication of the above is that future AGI will need to run on fast branching CPU _and_ massively parallel GPU with a fast data path between them - ie. balanced compute.
However, the best part will be discovering what the real Joanna writes about !