milkshakes

so far it's been working pretty well for me in most cases, but not always. having a few parallel efforts in motion means that even if one goes off the rails, i can pause it and still make progress on the others, then fix it up when i get home.

are they?

how do you define a user agent? is it desirable or not for users to be able to discover and access resources and communities on the internet using the tools, formats, and workflows that they prefer?

are search bots desirable?

the internet wayback machine?

how do you feel about browser extensions and greasemonkey scripts?

is the goal for your content to just be invisible to this growing population? is the expectation that all of this is just a fad, which will fade away? what is the end game to the tactics you have outlined? what is the strategy?

> One thing that should be learned from the bitter lesson is the great power of general purpose methods, of methods that continue to scale with increased computation even as the available computation becomes very great. The two methods that seem to scale arbitrarily in this way are search and learning.

in a street fight, the only rules are that there are no rules.

Moonshot AI Scale: Over 3.4 million exchanges

The operation targeted:

Agentic reasoning and tool use Coding and data analysis Computer-use agent development Computer vision Moonshot (Kimi models) employed hundreds of fraudulent accounts spanning multiple access pathways. Varied account types made the campaign harder to detect as a coordinated operation. We attributed the campaign through request metadata, which matched the public profiles of senior Moonshot staff. In a later phase, Moonshot used a more targeted approach, attempting to extract and reconstruct Claude’s reasoning traces.

if this were not the case, then we would be observing chinese models that far surpass frontier models in capabilities, rather than "almost as good, but much cheaper", and we would be having a very different conversation. what happens to these efforts when the subsidy is cut off?

and what will fund these budgets exactly? inference is cheap, distillation is cheap, training is what's expensive.

what is the end game for this strategy?

if the frontier labs shut down, or stop releasing to the public, and there's noting left to distill, how will you progress?

the system in place to prevent unauthorized abuse. by default, the guardrails are conservative. to reduce the guardrails you can jump through a progressive series of hoops to establish whether or not you have a valid use case. the entrypoint for establishing your use case is verifying your identity and background. if you don't want to do this, you are free to use Codex Security to identify and fix vulnerabilities, it is quite good at this. the harness and model are already evaluating the usage of the account and the nature of the code being examined and actions requested. but the again, the guardrail thresholds will be very conservative for anonymous users.

what is your proposal?

that's a bold statement. how does it not help solve the problem? what is a better solution?

https://projectzero.google/2021/12/a-deep-dive-into-nso-zero...

https://projectzero.google/2022/03/forcedentry-sandbox-escap...

exploiting vulnerabilities on hardened targets isn't just in a different league from finding them, it is a different sport altogether.

put simply, it's the difference between an integer overflow leading to a sandbox escaping RCE and one that leads to a crash.

Codex Security and 5.5/5.6 are still very good finding vulnerable code -- they will identify and fix unsafe behavior, but they will refuse to help you with exploitation -- they will actively prevent you from taking any steps to weaponize the unsafe behavior that are not required to remediate it. they will err conservative here, but for the most part they will still let you discover and address a wide range and depth of vulnerabilities. you can verify yourself to turn off the most basic safeguards and sign up through a more rigorous process for a spectrum of TAC options.

obviously there is a balance here -- openai wants to empower defenders while at the same time not exposing capabilities to the adversaries that would overwhelm defenders. there is no "right" answer. it is a work in progress. this is an intentional and deliberate decision to provide defenders with a (temporary, dwindling) advantage.

the example i chose was pretty extreme, but the underlying principle -- enable visibility discovery and remediation, but make it difficult to weaponize and defeat countermeasures makes sense given the bigger picture, IMO.

this calm before the storm is not going to last for very long, and defenders need every advantage they can get to get their houses in order before these capabilities are widely commoditized.