problem

We knew that US Big Tech had a problem with the costs of their AI. DeepSeek has just shown to what extent

by

DeepSeek is the new darling of AI. This family of models, developed by a Chinese R&D laboratory of the same name, has achieved what seemed impossible: compete with the OpenAI or Meta models and do so, according to them, at a much lower cost. Is that true? A development 18 times cheaper than GPT-4. The Chinese startup released DeepSeek V3 671B at the end of December 2024. Its gigantic model was trained in just two months with a budget of 5.58 million dollars according to SCMP and analysts cited in Financial Times. Its performance is comparable to OpenAI’s GPT-4, but the latter cost about $100 million to develop according to Sam Altman. That’s almost 18 times more if we take into account both the data revealed by SCMP and Altman’s estimates. Comparative cost of the main chat and reasoning models today. DeepSeek’s price is incredibly lower than its competitors. Data: DeepSeek, OpenAI, Anthropic, Meta. Amazingly cheap. The cost of DeepSeek’s API is incredibly low when compared to its competitors. If we take the data from DeepSeek, Goal, OpenAI, Google and Anthropic It seems to be clear that the cost of using DeepSeek through its API is much lower than that proposed by its rivals. We have included the cost of GPT-4o mini which seems to be the only one comparable, but its performance is much lower than DeepSeek V3. DeepSeek V3 is superior to most of its competitors, although it is true that Meta has released for example Llama 3.3 in recent days and that comparison varies frequently. And it is (theoretically) superior to all. As they point out on RedditDeepSeek V3 prices are promotional: starting February 8 they will be $0.27 per million input tokens (almost double) and $1.10 per million output tokens (almost four times more) . This makes the comparison somewhat better for the competitors, especially for Llama, the only one that can compete in cost although the Chinese model is superior to that of Meta (and almost also to the rest in many metrics) according to the benchmarks carried out in DeepSeek. DeepSeek also “thinks” cheaper. The cost comparison is not only in favor of DeepSeek in the area of ​​traditional chatbots, but also in the area of ​​reasoning models. According to its internal benchmarks, the spectacular DeepSeek R1 It is significantly superior to OpenAI’s o1, but using the o1 API costs 27 times more than that of DeepSeek R1. Hallucinatory. Price drop in sight. As expert Ethan Mollick points out, the market will adjust to these DeepSeek-driven price drops fairly quickly. According to their estimates, the cost of a GPT-4 level AI was reduced 1000 times in 18 months, and a 95% drop in the price of the reasoning models, which right now are clearly higher than the AI models behind ChatGPT, for example. a chinese tsunami. The launch of the DeepSeek models is a great little revolution for all types of developers of AI-based solutions: they now have access to much cheaper models that are comparatively equal to or superior to those of the competition. This puts their rivals in a lot of trouble, and we will see how they react. Good news for users. The truth is that for us, the users, as well as for the developers, this is great news, especially because these prices make access to these functions incredibly cheaper. The market has been following this trend clearly, but DeepSeek has made the jump in cost reduction suddenly drastic. Image | Xataka with Freepik Pikasso In Xataka | OpenAI prepares a PhD-level AI. It is so promising that he will first show it to the US Government

Malaysia came up with the idea of ​​painting its roads fluorescent to illuminate them at night. The problem was the price

by

In February 2024 we saw that the continuous line of a good part of the A-355, one of the roads with more accidents of Spanish territory, dawned with a thick continuous red line. It was a pilot measure by the DGT to highlight the prohibition of overtaking on that road. A few weeks earlier, a similar program was developed in Malaysia: repainting the road markings of certain roads with a photoluminescent paint. Not even a year has passed and it seems that they will not continue painting. The reason? It is too expensive, which has raised the question of… and they hadn’t thought of that before? The pilot. In mid-November 2023, the Government of Malaysia advertisement the launch of a pilot program that aimed to highlight the lines of certain roads. Using photoluminescent paint with the ability to glow at night. The authorities commented that these initiatives were suitable for dark areas that lacked public lighting. The lines shine for 10 hours and, in case of heavy rain, they also give off a certain shine that helps make them more visible, allowing drivers to stay in their lane. The images leave no room for doubt: it looks much better than the traditional white line, also better than reflective indicators. The project attracted attention. Initially, only 245 meters of road were painted at one intersection, representing about 490 meters of road markings, but other states began to carry out their own tests. Almost 20 times more expensive. The problems came practically from the beginning. The Malaysian Ministry of Works already warned that they were going to be very attentive to the results of the pilot program and the ability to paint on other roads, since the price of the new paint could be a problem. And so much so that it has ended up being an inconvenience. From the Ministry of Works reported that conventional paint costs about 40 ringgit per square meter, about 8.65 euros. Photoluminescent paint costs 749 ringgit per square meter, more than 160 euros. It costs 19 times more per square meter. Putting on the brake. Despite the enthusiasm with which the population received the measure, Malaysia’s deputy labor minister, Ahmad Maslan, commented a few weeks ago that it was not likely that they would continue painting. “The cost is too high, so we probably won’t continue with glow-in-the-dark lanes,” stated. It is curious when, months before, 31 roads had been identified that could have received this type of paint. Maslan stated that the tests did not meet the expectations of the ministry’s experts, without giving further explanations. Questions and suspicions. This is part of the eternal debate about priorities, budget and security, since we must remember that the areas that were going to receive this type of paintings are wooded and lack lighting, making it dangerous to travel through them due to interactions between humans. as by the action of animals crossing each other. And, evidently, the news has raised suspicions of corruption in local mediawith users who wondered if they didn’t know all this before starting the pilot project, what are those expectations that the painting has failed to meet… and that someone will have filled their pockets with the initial contract. Images | Alexander Nanta Linggi, Bernama (Azlim Mansor) In Xataka | The Tour de France has a team dedicated exclusively to something crucial: erasing penises painted on the roads

Trump has made it very clear that he wants to conquer Mars. Now NASA has the enormous problem of not being called SpaceX

by

Trump made just one space promise during his inauguration speech, but it was no small feat. The Martian dream. Between cheers and jumps of enthusiasm of Elon Musk, Donald Trump pointed out Mars as new “manifest destiny” of the United States. The newly inaugurated president promised to take astronauts to the Red Planet and plant the American flag in Martian soil. Trump stated: “We will pursue our manifest destiny to the stars, launching American astronauts to plant the flag on the planet Mars.” His words are not coincidental and have a clear influence, but they seem to mark a change of priorities for NASA that leaves the future of the Artemis lunar program. The influence of Elon Musk. “We’re going straight to Mars, the Moon is a distraction,” wrote the CEO of SpaceX two weeks ago. No one quite understood the scope of that tweet, since SpaceX has a very important contract with NASA to build the lander for the Artemis III and IV lunar missions, but now that message resonates on Capitol Hill. There, President Trump focused on Mars. It could just be a rhetorical statement (slowly, but surely, we have to get to Mars before China does), but with Elon Musk as a key ally, the new government could really be preparing a radical shift in astronautics strategy. The current Artemis program. It was precisely Trump’s first term that shaped NASA’s current lunar program. The then administrator, Jim Bridenstine, managed to put the United States’ return to the Moon on track with an architecture that combined NASA’s internal developments (the SLS rocket and the Orion spacecraft) with commercial spacecraft from private companies (SpaceX’s Starship HLS and the lunar module Blue Moon by Blue Origin). In turn, Bridenstine promoted a series of unmanned lunar missions and the creation of the Artemis Accordswhich already has 53 signatory countries, for international cooperation in future missions to the Moon, including the construction of a lunar base, the commercial exploitation of the satellite and everything that comes after (Mars, comets and asteroids). The Moon is a cruel lover. Artemis is not at her best. Manned missions have been delayed for problems on the Orion ship and delays in Starship development. Furthermore, the insane cost overruns of the SLS rocket have put a good part of public opinion against the current architecture of the program, which could be reconfigured with the support of SpaceX’s new Starship and Blue Origin’s New Glenn rockets. To make matters worse, the first two commercial missions associated with Artemis (CLPS‑1 and CLPS‑2) failed to reach the Moon or lie down when landing on the moonwhich has precipitated the cancellation of other more important missions such as NASA’s VIPER rover. But until Trump’s speech, there was nothing to predict that the Artemis program would be in danger. Is it really? From the Moon to Mars. Until now, NASA’s plan was to establish itself on the Moon throughout this decade and the next (or at least in the Gateway lunar station in orbit with the satellite) to prepare for the jump to Mars in the 2040s. Prioritize the Red Planet I would leave three scenarios to the foreseeable new administrator from NASA, Jared Isaacman: A reduced lunar program, without aspirations to create a large lunar base like the one proposed by the ILRS program led by China. Thus, the United States would continue in the race to put the first woman on the moon without stopping to focus on Mars. In exchange, he would cede lunar land to his opponents A bifurcated program with parallel lunar and Martian missions that do not throw away everything that has been developed so far. It would be the logical step if NASA’s budget were unlimited, but with the huge investment what the lunar program entails, adding a Martian program seems impossible A total redirection to the conquest of Mars. Following Elon Musk’s vision: the Moon is a distraction from the ultimate goal of become a multiplanetary civilization. Even with a majority in Congress, it is the option in which giants such as Boeing, Lockheed Martin, Northrop Grumman, and even New Space lose (Blue Origin has several lunar contracts). It seems complicated for congressmen to agree on a clean slate, but it is not totally impossible How would the United States get to Mars? There would be a public tender, but one option immediately comes to mind. NASA could adopt the SpaceX Mars program as your own. Elon Musk said SpaceX planned to launch five uncrewed Starships to Mars in 2026 and, if they managed to land, the first crewed mission to Mars in history in 2028. The experts agree at a crucial point: a mission with astronauts to Mars in the next four years is technically impossible if it is to be done with guarantees, since the scientific and technological challenges are monumental. But there were also many people convinced that Trump would not win the election again while Musk bet money that yes I would. Image | The White House, NASA In Xataka | Artemis has entered into crisis: NASA remains silent about the Orion spacecraft and rumors of cancellation of the SLS grow

Log In

Forgot password?

Forgot password?

Enter your account data and we will send you a link to reset your password.

Your password reset link appears to be invalid or expired.

Log in

Privacy Policy

Add to Collection

No Collections

Here you'll find all collections you've created before.