Good morning, and welcome to Tech News Now, TheStreet's daily tech rundown.
In today's edition, we're covering the death of the Apple Car, Google's efforts to pay publishers to test an unreleased AI platform, new information about the Tesla Roadster, the completion of SpaceX's Starship mishap investigation and Tumblr's decision to sell user data to Midjourney and OpenAI.
Tickers we're watching: (AAPL) , (TSLA) and (GOOG) .
DON'T MISS: OpenAI is seeking to dismiss several components of the New York Times' lawsuit against it, alleging that the Times paid someone to hack ChatGPT, something that a Times spokesperson dismissed as a bizarre mischaracterization.
It's a busy morning. Let's get into it.
Related: Analyst says Apple's latest move is a positive for Elon Musk's Tesla
The death of Project Titan
Bloomberg reported yesterday that after a decade of investment, Apple has decided to kill its Apple Car project, nicknamed Project Titan.
Many of the nearly 2,000 employees who made up the Apple Car team will be transferred to Apple's artificial intelligence team instead, underscoring the company's increasing behind-closed-doors focus on perfecting and shipping its own brand of generative AI.
Apple did not respond to TheStreet's request for comment regarding the report.
The move disappointed both Deepwater's Gene Munster and Wedbush's Dan Ives, who called the decision "another black eye" for the electric vehicle industry that underscores Tesla's dominance in the sector.
"While for Apple automotive is a bigger long-term revenue opportunity than generative AI, the decision to move on from Titan does have a benefit because it will increase the company's urgency to do something profound in generative AI," Munster said. "This is a positive for Tesla. I've long believed Apple was Tesla's biggest long-term US competitor."
Related: Apple could win the generative AI race — here's why it won't
Google is paying publishers to test unreleased AI tool
As questions and concerns of the copyright infringement seemingly integral to generative AI continue to plague the industry, Google, according to AdWeek, is paying news publishers to beta test a private generative AI platform.
Google did not immediately respond to TheStreet's request for comment.
The program — which pays publishers a monthly stipend that amounts to a five-figure annual payment — requires publishers to use the platform to publish three articles a day, one newsletter each week and one marketing campaign per month, for 12 months.
It is unclear which publishers — and how many of them — are participating in the program.
You may consider:
— Jason Kint (@jason_kint) February 27, 2024
(a) why so secretive?
(b) why is Google paying for use of something valued by publishers? Please see (a).
(c) color code plagiarism. thanks.
(d) what Google may learn from paying humans to edit generative AI reports scraping OPP (other people's property).
-bye- https://t.co/jsr8zuyu4I
The program, according to the report, functions essentially as an AI-powered aggregator; publishers using the program list the websites of other news organizations that publish relevant content — whenever one of those websites publishes an article, that article appears on the program's dashboard where it is then re-worded by the generative AI tool.
The external websites whose content is being scraped were not asked their permission; likewise, the AI-produced aggregate articles are not required to be labeled as being produced using generative AI tools.
“In partnership with news publishers, especially smaller publishers, we’re in the early stages of exploring ideas to potentially provide AI-enabled tools to help journalists with their work,” a Google representative told AdWeek. “These tools are not intended to, and cannot, replace the essential role journalists have in reporting, creating and fact-checking their articles.”
Related: Human creativity persists in the era of generative AI
Musk: New Tesla Roadster goes 0-60 in less than 1 second
Another long-delayed Tesla model is finally close to seeing the light of day, according to CEO Elon Musk. Tesla is aiming to ship its new Roadster, which was initially scheduled to be launched in 2020, next year, according to Musk.
He said that the production design of the car will be unveiled by the end of this year.
You will love the new Roadster more than your house
— Elon Musk (@elonmusk) February 28, 2024
"Tonight, we radically increased the design goals for the new Tesla Roadster," Musk wrote in a post on X. "There will never be another car like this, if you could even call it a car."
He added that the car will be able to go from 0 to 60 mph in less than one second, saying: "And that is the least interesting part."
Such a feat is incredibly rare, but not impossible: Swiss students in September purpose-built a 309-pound electric car that accelerated from rest to 62 mph in .956 seconds.
Related: Here's why the Tesla bears are starting to outnumber the bulls
Starship is one step closer to a third launch
The Federal Aviation Administration (FAA) announced Feb. 26 that it has closed its mishap investigation into SpaceX's November Starship launch, in which the rocket flew for nearly eight minutes before exploding.
SpaceX identified the root causes of the failure and 17 corrective actions, which have been accepted by the FAA.
Among these causes and fixes was an issue with oxygen venting which Musk in January said triggered a fire and caused an explosion.
Before it is able to fly Starship a third time, SpaceX must implement each of these corrective actions, then apply for a license modification from the FAA.
The FAA has not yet approved a third launch of Starship.
"More Starships are ready to fly, putting flight hardware in a flight environment to learn as quickly as possible," SpaceX said in a stateme2nt. "Recursive improvement is essential as we work to build a fully reusable launch system capable of carrying satellites, payloads, crew, and cargo to a variety of orbits and Earth, lunar, or Martian landing sites."
Related: The critical differences between Elon Musk's first and second Starship flight
The AI Corner: Tumblr and WordPress to sell user data to AI companies
Tumblr and WordPress.com are planning to sell user data to OpenAI and Midjourney, according to a recent 404 Media report, which highlighted internal documents and a source with knowledge of the deal.
Neither Tumblr, nor Automattic, the company behind WordPress.com, Tumblr, Jetpack and several other brands, responded to a request for comment.
The deals, according to the report, are imminent.
It remains unclear exactly what data is a part of these deals, though internal documents showed that Tumblr has already prepared an initial data dump that included a large number of user posts it wasn't supposed to include.
After the report was published, Automattic released a statement saying that users can opt out of sharing of their content.
"We are also working directly with select AI companies as long as their plans align with what our community cares about: attribution, opt-outs, and control," the company said. "Our partnerships will respect all opt-out settings. We also plan to take that a step further and regularly update any partners about people who newly opt out and ask that their content be removed from past sources and future training."
More deep dives on AI:
- ChatGPT maker says 'it would be impossible' to train models without violating copyright
- Copyright expert predicts result of NY Times lawsuit against Microsoft, OpenAI
- Meta and IBM team up against dominant Big Tech players
It remains unclear if self-hosted WordPress sites will be impacted by the coming deal.
According to the report, internal documents showed that there is no guarantee that partners will honor opt-out requests, though Automattic's head of AI Andrew Spittle believes "partners will honor this based on our conversations with them to this point."
This latest almost-deal comes as tech companies are pursuing data at all costs, copyright be damned. Both OpenAI and Midjourney are embroiled in multiple lawsuits alleging rampant copyright infringement in both the input and output of their AI models.
OpenAI has argued regularly that it is fair to train its highly commercialized AI models on content with permission, compensation or credit, though the company is also pursuing licensing deals with publishers. It is a stance that the humans who create said content largely disagree with.
OpenAI said recently that "it would be impossible to train today's leading AI models without using copyrighted materials."
Contact Ian with AI stories via email, ian.krietzberg@thearenagroup.net, or Signal 732-804-1223.
Related: How Uniphore is expanding control and reducing hallucinations within its AI models
