I heard the other day that the age of cloud-based AI was returning us back to thin clients. Instead, we get the worst of both worlds.
(Not my Mac, obviously. Spotted on Zuck’s rubbish ragebait website.)
| website | https://tommorris.org/ |
I heard the other day that the age of cloud-based AI was returning us back to thin clients. Instead, we get the worst of both worlds.
(Not my Mac, obviously. Spotted on Zuck’s rubbish ragebait website.)
Move over “too big to fail”, welcome “too American to regulate”.
Podcast advertising is wild.
Guest: "...and that's how generative AI has been used by neofascists to create propaganda to elect governments that will act in cruel and selfish ways that ultimately undermine any chance of us preserving human life in a livable environment, but don't worry too much about that, we'll probably end up completely obliterating each other with nukes based on the most ridiculous social media posts imaginable."
Ad: "get 10% off a mattress now with code ESCHATON10"
Obviously the system is two-tier. The mistake is people who think those who are in charge of it are the type of person running a diversity scheme in a corporate HR department in Swansea rather than those with enormous amounts of wealth and political power.
Janet in HR handing out rainbow diversity badges didn’t sell the country off to private equity vultures. The person who did is super keen on you being angrier with Janet than the people with actual power.
Every week someone seems keen on arguing that there is “two-tier” justice in Britain.
Of course there is. If you’re a Royal or a chum of Epstein (or both), you won’t be charged.
If you cause the largest miscarriage of justice in English history (Post Office scandal), you likely won’t be charged.
If you go to Oxford and/or Eton, you massively reduce any risk of criminal charge.
If you are an MP, you can make false promises and gain a material benefit. In other walks of life, that’s fraud.
Also: "it is possible for Company X to do Y" does not mean "Company X IS DOING Y".
Nor does "Company X said they are going to do Y" + "they later said they would not do Y" + "Company X has lied about things in the past" = "Company X is definitely doing X and lying about it".
Tech companies suck obvs, but paranoia doesn't help. There are a whole series of practical, evidence-based (often free) steps you can take to protect yourself and others. An evidence-free counsel of despair hinders this.
Similarly, if you are trying to answer "between Google, Apple and Microsoft, who is the most/least trustworthy on privacy/security?" you need to have meaningful metrics on judging this, and a rough idea of the kind of user (i.e. what their threat model is).
An Edward Snowden or Chelsea Manning (or an MP or human rights activist) has different privacy/security problems from an ordinary person.
If someone doesn't make those assumptions explicit, their conclusion might not be worth much.
Saw a completely wild video this morning on privacy/security from a popular YouTuber. I won't say who (got better things to do than getting sued for libel) but let's just say this...
I really don't like non-technical people being stirred up through half-truths into paranoia on topics like privacy/security (in the vein of "your phone is always listening to you!!1!"). Yes, don't trust big tech, but also just because someone at Google/Apple/MS said the sky is blue doesn't mean it is false.
Life is hard for Copilot users.
First, you spend 50% of the monthly token budget generating a React component or a shit hello world bash script that doesn't work.
Then you spend the rest generating a Reddit post to express your anger about how many tokens you've used.
There is another alternative available, namely just learning to code using your brain. No corporation in the world can take that away from you.