This text belongs to ‘Artificial’, the newsletter on AI that Delia RodrÃguez sends out every Friday. If you want to receive it, sign up here.
Dear readers, dear readers:
Let’s start with a very boring and very interesting subject at the same time.
A little over a month ago, students from the Robotics and Artificial Intelligence Club at IE University of Madrid filled out the web form where OpenAI, the creator of ChatGPT, appealed to those who wanted to meet with them during their 2023 Tour. They were also contacted by a school teacher, Joe Haslam. The company responded to this professor by accepting the invitation and, well, they were all very happy.
The following was an act, held last Monday at the offices of the business school, where Sam Altman, founder of the company, spoke to 600 teachers, students, entrepreneurs, investors and other guests. He also met with the board of directors of IndesIA, the consortium of large Spanish companies for AI. There were cakes to attend. Anyone who wants to gossip who was, who wasn’t and what was said that afternoon in one of the Castellana towers with the man of the moment, the right place is LinkedIn.
The talk began with Altman explaining that meeting with politicians was new to him.
On Monday he was with Sánchez in Madrid.
On Tuesday with Morawiecki in Warsaw and Macron in Paris.
On Wednesday he met Sunat in London. In one act he threatened that if OpenAI could not comply with European legislation he would leave the continent.
On Thursday he went to Munich. The European commissioner for the internal market, Thierry Breton, sent him to hell saying that he does not haggle with European laws.
Altman is not the only one lobbying Europe begging for the mess he is about to create to be dealt with. With Google CEO Sundai Pichar, the commissioner was a bit kinder. They sat down to try and get a quick deal with Europe pending the European AI regulation.
The two CEOs have not only lobbied in person this week, but also in writing. In a short post on its blog, OpenAI reiterated the position it defended before the US Senate: an international regulatory body is necessary, and one thing is the control of large models and another, the freedom that smaller and open source projects need. . Pichar repeated on the FT that AI is the deepest technology humanity is working on right now, and that building it responsibly is the only race that matters.
The details, of course, are something else. It seems that the only thing that everyone is clear about is that this must be legislated and quickly. Will Europe and the United States be able to reach an agreement in a reasonable period of time to agree on a regulation with common points, when they never succeed in technological matters?
But back to this week’s star, Sam Altman. Who is it? According to the profiles and interviews we have read about him:
What else has happened this week
– Apple and other large companies, including some Spanish ones, prohibit the use of ChatGPT. It seems sensible, considering that the data that is entered is used to further improve the system itself.
– Microsoft has flooded its annual developer conference with announcements about integrating AI into its products. Copilot, the Windows assistant, will work like a chat where you can ask whatever comes to your mind: find a trip, summarize this PDF, play music, send this. Once I have seen the presentation video, I no longer understand how computers could have worked in another way. Just add the voice and we would be in the future: one in which anyone can talk to machines in their own words. I’m really intrigued to see what Apple will present. “Whoever wins the personal agent, that’s the big thing, because you’ll never go to a search site again, you’ll never go to a productivity site again, you’ll never go to Amazon again,” Bill Gates has said very clearly.
Microsoft’s vision is to make it easy for developers to create co-pilots, connected with a broad ecosystem of plug-ins that will allow them to do almost anything. On this, it is interesting to read this profile on the CTO of Bing and promoter of the Microsoft-OpenAI agreement, Kevin Scott.
For example, through Azure, Microsoft will make it easier for companies to create corporate co-pilots with access to internal documentation. If this succeeds in making intranets obsolete, the extinction of humanity will have compensated.
Also watch out for its web page construction tool, to which you can give direct orders. Its design tools with Dall-E (Bing Creator and Designer) will incorporate a cryptographic watermark, something demanded by professionals.
– Adobe has launched Generative Fill in beta, a new Photoshop functionality thanks to which you open the program, surround the area of ​​the image you want to edit, tell the AI ​​what you want to do and it executes. Ramón Peco is testing it for La Vanguardia, but in the meantime I highly recommend watching the video because it seems like magic.
– Priests also use AI. Report by MarÃa-Paz López from Berlin.
– “The machine does not understand what it is doing, and this also happens to our brain. The mechanisms that prepare a sentence –or any mental phenomenon– are not conscious. The brain does not know the sentence it is forming. We are aware of it at the moment of enunciation. That is why E.M. Forster was right when he made one of his characters say: how will I know what I think, if I haven’t said it yet? You should read this article by José Antonio Marina on how thought-machine and thought-human are similar.
– Kissinger turns 100 tomorrow, and believes that to avoid a world war, the US and China have no choice but to understand each other and that they have between five and ten years to do so. “But we are not in a normal circumstance,” he says in this article from The Economist translated in La Vanguardia, “because of mutual assured destruction and artificial intelligence.”
– Javier Pastor remembers the metaphor of the bicycle and tries SudoWrite, the tool that worries those of us who write.
iAnxiety Levels This Week: Elevated, but more from reading Kissinger than the Altman tour. The feeling is tempered because this week I actually feel a little stupid. A collaborator of this newspaper sent me an audio that I forwarded to LuzÃa -the WhatsApp assistant that I recommended last week- so that she could transcribe it because she was doing several things at the same time. The text ended with a question that I answered, without realizing that I was not speaking with the collaborator, but with LuzÃa, who answered me in turn. And so we had an animated work talk in which he promised to write an article for the next day. I only realized the error because no one uses the exclamation mark at the beginning of a sentence anymore.
