Re: quarantines for fully vaccinated travellers
The problem a national government has in setting quarantine standards for vaccinated people is that there are so many inconsistencies to vaccines, and the topic is wrapped in geopolitics.
Do you exempt (from quarantine) those who claim to have gotten vaccines in countries where vaccine documentation is easy to fake? What about countries where vaccine documentation is nonstandard between regions?
How do you (cheaply) verify vaccine documentation from each foreign country / region?
Does refusing to exempt people from poorly-documented countries cause a diplomatic incident with those countries?
Do you exempt those who have documented vaccines which are not authorized in your country? If so, what are your criteria for accepting vaccines?
If not, then the home countries of inelligible vaccines will lodge a diplomatic protest, and maybe accuse you of racism.
Or do you only exempt those who got vaccines which are produced in your country? (Chinese-style vaccine nationalism.)
Do you exempt those who got vaccines which are authorized in your country, but are coming from countries where immunity-escaping strains are common?
Do you exempt those who got the vaccine, but are in a demographic likely to have poor immune response?
If you exempt people with vaccine immunity, do you exempt people with positive COVID antibody tests?
If you exempt people with antibody tests, which tests do you exempt? See all the previous questions about reliability of documentation again...
The cost to administration of such an exemption program is high, the social, political, and medical perils are many, and there is the chance of quarantine escape in >5% of cases. Vaccine exemption from national quarantine is unlikely.
I’m happy that this was done before release. However … I’m still left wondering “how many prompts did they try?” In practice, the first AI self-replicating escape is not likely to be a model working alone on a server, but a model carefully and iteratively prompted, with overall strategy provided by a malicious human programmer. Also, one wonders what will happen once the base architecture is in the training set. One need only recognize that there is a lot of profit to be made (and more cheaply) by having the AI identify and exploit zero-days to generate and spread malware (say, while shorting the stock of a target company). Perhaps GPT-4 is not yet capable enough to find or exploit zero-days. I suppose we will find out soon enough.
Note that this creates a strong argument for never open-sourcing the model once a certain level of capability is reached: a GPT-N with enough hints about its own structure will be able to capably write itself.