While Block’s Jack Dorsey lays off thousands because of AI, with market approval, the announcement from Anthropic’s head that he will not let AI control weapons is more nuanced than it seems. The objection is technological, not moral.
Profit does not stop redundancies; Wall Street rewards the cuts
4,000 people found out on X1Directly from x.com/jack that they are to be made redundant because of AI, even as the boss boasts that profits have risen. Either Block CEO Jack Dorsey is a poor employer who filled his offices with too many people (and, horror of horrors, Elon Musk was right), or last night’s announcement is a sign of the world to come.
Anyone who looked more closely at Block, Dorsey’s company, saw that many hires were made during the pandemic when share values soared, but for several years these have returned to pre-2020 levels; so, perhaps there is something else at play.
On the other hand, the way Wall Street reacted (shares rose by 24%)2 Block shares soar 24% as company slashes workforce by nearly half, cnbc.com says a lot about how employees are viewed through spreadsheets. And about how future decisions will be made when the focus is placed firmly on the figures.
The truth behind Anthropic’s refusal of the Pentagon
In other news, Anthropic has announced3 Statement from Dario Amodei on our discussions with the Department of War, anthropic.com that it is not coming to terms with the Pentagon, and that it will not let Claude control weapons of its own accord or be used in the mass surveillance of citizens. OpenAI and Google employees have also signed a petition to this effect.4 We Will Not Be Divided, notdivided.org
However, if you read Dario Amodei’s statement more closely – one of the proponents of the idea that an AI development moratorium is impossible because others will not stop – you will see that the objection is rather because the Pentagon’s demands are ” simply outside the bounds of what today’s technology can safely and reliably do.”
Anthropic will continue to work with the Department of War; it has absolutely no problem with the use of mass surveillance in non-domestic contexts (and carefully avoids saying what happens if it were used in the surveillance of those who are not citizens, perhaps also because it is not within this department’s remit). Amodei has no problem with autonomous weapons either, in a future where they can be developed safely. The news is not as positive as one might think.