271 karma · joined March 4, 2024
https://www.anthropic.com/threat-intelligence-report-septemb...
> Biological misuse is one of the most serious risks of frontier AI models. It has long been a concern that AI models might one day reach the level of capability where they can help to make existing pathogens more dangerous—or create entirely new ones. Without the correct safeguards, such capabilities could have catastrophic consequences.
Results from evaluations of older models (for example Claude Opus 4 and Claude Sonnet 4.5, from 2025) clearly showed that these models were well below the threshold where they could meaningfully assist a sophisticated user in carrying out dangerous biological research. As a result, safeguards on these models were less stringent, directed mostly at preventing access to content that might uplift novices in recreating known bioweapons. But for today’s models—which are capable of assisting in a range of complex scientific research tasks—the evidence is no longer certain, and we cannot make that same assurance. For this reason, and out of an abundance of caution, we have launched recent models (most notably Claude Fable 5) with stronger safeguards that restrict access to a wide range of dual-use biological research queries.
...
Here, we present five case studies of actors using our models in ways that could support biological weapons development. These examples are illustrative of the kinds of tasks to which our models are put, and the often-difficult judgements we have to make when assessing whether or not a given biological use is dangerous. They also convey that we encounter what would otherwise be non-public insight into the risks associated with biological misuse from AI
Please spend even 1-2 minutes on the https://livingwage.mit.edu/ for example this calculation for Philadelphia: https://livingwage.mit.edu/metros/37980
"PocketBase is an open source backend consisting of embedded database (SQLite) with realtime subscriptions, builtin auth management, convenient dashboard UI and simple REST-ish API. It can be used both as Go framework and as standalone application.
The easiest way to get started is to download the prebuilt minimal PocketBase executable"
Edit: I missed the point here, see reply.
What a contribution!
Jokes aside, if you support gun control you should think very seriously about if you believe a company like this should be legally allowed to do what it's doing. I personally do not.
I've seen people who work on art nonstop and think their art is really meaningful, get really into it, but there's little output that anyone cares about or that changes the world or that anyone would pay for.
On the other hand I've had friends who created some art, and have real customers paying real money, ....so it's easy to tell their art project is meaningful. (is that the conclusion?)
Maybe something (AI, code, an AI-coded project) doesn't need to have value to others or be worth money for it to be meaningful to the person who created it.
Coaster brakes are an example of a simpler design, that requires less maintenance, that are cheaper, that are poor in some environments like hills but are fine in flat areas.
Some would rather walk their bike down a hill but have a bike that never needs brake maintenance than to have better brakes that are a point of failure.
The same drive that pushes people to trying to build offgrid homes is slowly growing in spaces of material goods and home goods and such. Add environmentalism to it also.
Or if you see AI as more tool and less entity, better gunsafes for our guns.
The model doesn't have preferences, the model expresses the most likely preferences that the training data would have expressed (humans).
This feels like throwing a tennis ball at a brick wall then interpreting the angle and direction that the ball travels as it bounces off as the tennis ball or wall's preferred direction of travel.
You are the one who threw the ball. You are measuring your own preferences.
Sounds worse than it is since it's progressive taxation meaning the effective tax rate for the entire 60k EUR is somewhere like ~29%, not 45.
€0 to €12,450: 19%
€12,451 to €20,200: 24%
€20,201 to €35,200: 30%
€35,201 to €60,000: 37%
Source: Gemini so it could be complete bullshit.
To be fair, VAT is heaver in Europe than for example US, so the overall tax burden might still feel more onerous.
https://www.energy-concepts.com/_pages/app_isaac_solar_ice_m...
I would be curious what you think of the design.
1) what benefits are there to my AI's memory having the property of being able to prove, cryptographically, what was stored and when? My first reaction is that this sounds like something that is probably expensive to achieve but that I don't intuitively grok the benefits of.
2) why should I allow you to decide the type system of my knowledge, and how flexible is your system to the possibility that you've got it wrong and this typeset is not the right design?: "Types include: fact (stable truths), state (mutable context), event (dated milestones), procedure (proven workflows and corrections), relation (links between concepts), domain, and task."
3) RE: "Session bootstrap is a single MCP tool call — memory_session_bootstrap — that loads relevant atoms, active procedures, Markov predictions, and conflicting facts at the start of every session. You pass your objective and it returns a pre-ranked, token-budgeted context block. Bootstrap uses both Markov arcs and knowledge graph edges for scoring" - it seems like bootstrapping sessions can be a very preference-based, personalized thing, in other words - different developers have a different opinion about what matters. Unless you can prove that your bootstrap methodology achieves some higher benchmark than any random person off the street's bootstrap technique/instructions to an agent, then I find it hard to trust this and want to pay for it over just writing my own agent instructions and triggering them with a CLI command or alias.
Under another set of criteria (environmental concerns), it's probably a positive.
Many many folks are the type that is willing to hard grind/suffer short-term to get through a hoop, but as soon as they are inside they turn that 'optimizer' mindset towards 'how can I do the minimum necessary to coast and collect my paycheck'.
And many many folks who are highly motivated to work hard every day at their job are not highly motivated to prepare for jumping through a hoop like a circus clown.
It's that if you don't have a way to measure quality yourself (you're not an expert), or if you don't ask for exact specific features/qualities even if you offer a high budget and ask for high quality, manufacturers absolutely will cut corners, do stupid things, etc. to increase their own margin at the expense of quality and just hope you are none the wiser.
The US had the same culture in certain periods and certain industries, so I am not suggesting it's some innate Chinese characteristic. The meat factories exposed in Upton Sinclair's The Jungle, where he documented unsanitary conditions, spoiled meat, rat infestations etc come to mind.
But one cannot just ignore that the China of today has had problems with fake baby formula, fake alcohol, fake medicine, fake everything and expect that that doesn't make its way into manufacturing of other classes of goods.
Help me prioritize what matters most.
If someone asks me a question, research the answer as deeply as you can and draft a reply for me, but do not send it."
This is a very dangerous road to go down. You may feel like you are getting more done but end up living your life on autopilot, without any introspection or applying your own taste.