Technology

Why Silicon Valley is so excited about awkward drawings done by artificial intelligence

Published

3 years ago

October 9, 2022

admin

Stable Diffusion’s web interface, DreamStudio

Screenshot/Stable Diffusion

Computer programs can now create never-before-seen images in seconds.

Feed one of these programs some words, and it will usually spit out a picture that actually matches the description, no matter how bizarre.

The pictures aren’t perfect. They often feature hands with extra fingers or digits that bend and curve unnaturally. Image generators have issues with text, coming up with nonsensical signs or making up their own alphabet.

But these image-generating programs — which look like toys today — could be the start of a big wave in technology. Technologists call them generative models, or generative AI.

“In the last three months, the words ‘generative AI’ went from, ‘no one even discussed this’ to the buzzword du jour,” said David Beisel, a venture capitalist at NextView Ventures.

In the past year, generative AI has gotten so much better that it’s inspired people to leave their jobs, start new companies and dream about a future where artificial intelligence could power a new generation of tech giants.

The field of artificial intelligence has been having a boom phase for the past half-decade or so, but most of those advancements have been related to making sense of existing data. AI models have quickly grown efficient enough to recognize whether there’s a cat in a photo you just took on your phone and reliable enough to power results from a Google search engine billions of times per day.

But generative AI models can produce something entirely new that wasn’t there before — in other words, they’re creating, not just analyzing.

“The impressive part, even for me, is that it’s able to compose new stuff,” said Boris Dayma, creator of the Craiyon generative AI. “It’s not just creating old images, it’s new things that can be completely different to what it’s seen before.”

Sequoia Capital — historically the most successful venture capital firm in the history of the industry, with early bets on companies like Apple and Google — says in a blog post on its website that “Generative AI has the potential to generate trillions of dollars of economic value.” The VC firm predicts that generative AI could change every industry that requires humans to create original work, from gaming to advertising to law.

In a twist, Sequoia also notes in the post that the message was partially written by GPT-3, a generative AI that produces text.

How generative AI works

Kif Leswing/Craiyon

Image generation uses techniques from a subset of machine learning called deep learning, which has driven most of the advancements in the field of artificial intelligence since a landmark 2012 paper about image classification ignited renewed interest in the technology.

Deep learning uses models trained on large sets of data until the program understands relationships in that data. Then the model can be used for applications, like identifying if a picture has a dog in it, or translating text.

Image generators work by turning this process on its head. Instead of translating from English to French, for example, they translate an English phrase into an image. They usually have two main parts, one that processes the initial phrase, and the second that turns that data into an image.

The first wave of generative AIs was based on an approach called GAN, which stands for generative adversarial networks. GANs were famously used in a tool that generates photos of people who don’t exist. Essentially, they work by having two AI models compete against each other to better create an image that fits with a goal.

Newer approaches generally use transformers, which were first described in a 2017 Google paper. It’s an emerging technique that can take advantage of bigger datasets that can cost millions of dollars to train.

The first image generator to gain a lot of attention was DALL-E, a program announced in 2021 by OpenAI, a well-funded startup in Silicon Valley. OpenAI released a more powerful version this year.

“With DALL-E 2, that’s really the moment when when sort of we crossed the uncanny valley,” said Christian Cantrell, a developer focusing on generative AI.

Another commonly used AI-based image generator is Craiyon, formerly known as Dall-E Mini, which is available on the web. Users can type in a phrase and see it illustrated in minutes in their browser.

Since launching in July 2021, it’s now generating about 10 million images a day, adding up to 1 billion images that have never existed before, according to Dayma. He’s made Craiyon his full-time job after usage skyrocketed earlier this year. He says he’s focused on using advertising to keep the website free to users because the site’s server costs are high.

A Twitter account dedicated to the weirdest and most creative images on Craiyon has over 1 million followers, and regularly serves up images of increasingly improbable or absurd scenes. For example: An Italian sink with a tap that dispenses marinara sauce or Minions fighting in the Vietnam War.

But the program that has inspired the most tinkering is Stable Diffusion, which was released to the public in August. The code for it is available on GitHub and can be run on computers, not just in the cloud or through a programming interface. That has inspired users to tweak the program’s code for their own purposes, or build on top of it.

For example, Stable Diffusion was integrated into Adobe Photoshop through a plug-in, allowing users to generate backgrounds and other parts of images that they can then directly manipulate inside the application using layers and other Photoshop tools, turning generative AI from something that produces finished images into a tool that can be used by professionals.

“I wanted to meet creative professionals where they were and I wanted to empower them to bring AI into their workflows, not blow up their workflows,” said Cantrell, developer of the plug-in.

Cantrell, who was a 20-year Adobe veteran before leaving his job this year to focus on generative AI, says the plug-in has been downloaded tens of thousands of times. Artists tell him they use it in myriad ways that he couldn’t have anticipated, such as animating Godzilla or creating pictures of Spider-Man in any pose the artist could imagine.

“Usually, you start from inspiration, right? You’re looking at mood boards, those kinds of things,” Cantrell said. “So my initial plan with the first version, let’s get past the blank canvas problem, you type in what you’re thinking, just describe what you’re thinking and then I’ll show you some stuff, right?”

An emerging art to working with generative AIs is how to frame the “prompt,” or string of words that lead to the image. A search engine called Lexica catalogs Stable Diffusion images and the exact string of words that can be used to generate them.

Guides have popped up on Reddit and Discord describing tricks that people have discovered to dial in the kind of picture they want.

Startups, cloud providers, and chip makers could thrive

Image generated by DALL-E with prompt: A cat on sitting on the moon, in the style of Pablo Picasso, detailed, stars

Screenshot/OpenAI

Some investors are looking at generative AI as a potentially transformative platform shift, like the smartphone or the early days of the web. These kinds of shifts greatly expand the total addressable market of people who might be able to use the technology, moving from a few dedicated nerds to business professionals — and eventually everyone else.

“It’s not as though AI hadn’t been around before this — and it wasn’t like we hadn’t had mobile before 2007,” said Beisel, the seed investor. “But it’s like this moment where it just kind of all comes together. That real people, like end-user consumers, can experiment and see something that’s different than it was before.”

Cantrell sees generative machine learning as akin to an even more foundational technology: the database. Originally pioneered by companies like Oracle in the 1970s as a way to store and organize discrete bits of information in clearly delineated rows and columns — think of an enormous Excel spreadsheet, databases have been re-envisioned to store every type of data for every conceivable type of computing application from the web to mobile.

“Machine learning is kind of like databases, where databases were a huge unlock for web apps. Almost every app you or I have ever used in our lives is on top of a database,” Cantrell said. “Nobody cares how the database works, they just know how to use it.”

Michael Dempsey, managing partner at Compound VC, says moments where technologies previously limited to labs break into the mainstream are “very rare” and attract a lot of attention from venture investors, who like to make bets on fields that could be huge. Still, he warns that this moment in generative AI might end up being a “curiosity phase” closer to the peak of a hype cycle. And companies founded during this era could fail because they don’t focus on specific uses that businesses or consumers would pay for.

Others in the field believe that startups pioneering these technologies today could eventually challenge the software giants that currently dominate the artificial intelligence space, including Google, Facebook parent Meta and Microsoft, paving the way for the next generation of tech giants.

“There’s going to be a bunch of trillion-dollar companies — a whole generation of startups who are going to build on this new way of doing technologies,” said Clement Delangue, the CEO of Hugging Face, a developer platform like GitHub that hosts pre-trained models, including those for Craiyon and Stable Diffusion. Its goal is to make AI technology easier for programmers to build on.

Some of these firms are already sporting significant investment.

Hugging Face was valued at $2 billion after raising money earlier this year from investors including Lux Capital and Sequoia; and OpenAI, the most prominent startup in the field, has received over $1 billion in funding from Microsoft and Khosla Ventures.

Meanwhile, Stability AI, the maker of Stable Diffusion, is in talks to raise venture funding at a valuation of as much as $1 billion, according to Forbes. A representative for Stability AI declined to comment.

Cloud providers like Amazon, Microsoft and Google could also benefit because generative AI can be very computationally intensive.

Meta and Google have hired some of the most prominent talent in the field in hopes that advances might be able to be integrated into company products. In September, Meta announced an AI program called “Make-A-Video” that takes the technology one step farther by generating videos, not just images.

“This is pretty amazing progress,” Meta CEO Mark Zuckerberg said in a post on his Facebook page. “It’s much harder to generate video than photos because beyond correctly generating each pixel, the system also has to predict how they’ll change over time.”

On Wednesday, Google matched Meta and announced and released code for a program called Phenaki that also does text to video, and can generate minutes of footage.

The boom could also bolster chipmakers like Nvidia, AMD and Intel, which make the kind of advanced graphics processors that are ideal for training and deploying AI models.

At a conference last week, Nvidia CEO Jensen Huang highlighted generative AI as a key use for the company’s newest chips, saying these kind of programs could soon “revolutionize communications.”

Profitable end uses for Generative AI are currently rare. A lot of today’s excitement revolves around free or low-cost experimentation. For example, some writers have been experimented with using image generators to make images for articles.

One example of Nvidia’s work is the use of a model to generate new 3D images of people, animals, vehicles or furniture that can populate a virtual game world.

Ethical issues

Prompt: “A cat sitting on the moon, in the style of picasso, detailed”

Screenshot/Craiyon

Ultimately, everyone developing generative AI will have to grapple with some of the ethical issues that come up from image generators.

First, there’s the jobs question. Even though many programs require a powerful graphics processor, computer-generated content is still going to be far less expensive than the work of a professional illustrator, which can cost hundreds of dollars per hour.

That could spell trouble for artists, video producers and other people whose job it is to generate creative work. For example, a person whose job is choosing images for a pitch deck or creating marketing materials could be replaced by a computer program very shortly.

“It turns out, machine-learning models are probably going to start being orders of magnitude better and faster and cheaper than that person,” said Compound VC’s Dempsey.

There are also complicated questions around originality and ownership.

Generative AIs are trained on huge amounts of images, and it’s still being debated in the field and in courts whether the creators of the original images have any copyright claims on images generated to be in the original creator’s style.

One artist won an art competition in Colorado using an image largely created by a generative AI called MidJourney, although he said in interviews after he won that he processed the image after choosing it from one of hundreds he generated and then tweaking it in Photoshop.

Some images generated by Stable Diffusion seem to have watermarks, suggesting that a part of the original datasets were copyrighted. Some prompt guides recommend using specific living artists’ names in prompts in order to get better results that mimic the style of that artist.

Last month, Getty Images banned users from uploading generative AI images into its stock image database, because it was concerned about legal challenges around copyright.

Image generators can also be used to create new images of trademarked characters or objects, such as the Minions, Marvel characters or the throne from Game of Thrones.

As image-generating software gets better, it also has the potential to be able to fool users into believing false information or to display images or videos of events that never happened.

Developers also have to grapple with the possibility that models trained on large amounts of data may have biases related to gender, race or culture included in the data, which can lead to the model displaying that bias in its output. For its part, Hugging Face, the model-sharing website, publishes materials such as an ethics newsletter and holds talks about responsible development in the AI field.

“What we’re seeing with these models is one of the short-term and existing challenges is that because they’re probabilistic models, trained on large datasets, they tend to encode a lot of biases,” Delangue said, offering an example of a generative AI drawing a picture of a “software engineer” as a white man.

Technology

Tesla Optimus robotics vice president Milan Kovac is leaving the company

Published

3 hours ago

June 6, 2025

admin

Tesla Optimus robotics vice president Milan Kovac is leaving the company

Tesla displays Optimus next to two of its vehicles at the World Robot Conference in Beijing on Aug. 22, 2024.

CNBC | Evelyn

Tesla’s vice president of Optimus robotics, Milan Kovac, said on Friday that he’s leaving the company.

In a post on X, Kovac thanked Tesla CEO Elon Musk and reminisced about his tenure, which began in 2016.

“I want to thank @elonmusk from the bottom of my heart for his trust and teachings over the decade we’ve worked together,” Kovac wrote. “Elon, you’ve taught me to discern signal from noise, hardcore resilience, and many fundamental principles of engineering. I am forever grateful. Tesla will win, I guarantee you that.”

Tesla is developing Optimus with the aim of someday selling it as a bipedal, intelligent robot capable of everything from factory work to babysitting.

In a first-quarter shareholder deck, Tesla said it was on target for “builds of Optimus on our Fremont pilot production line in 2025, with wider deployment of bots doing useful work across our factories.”

During Tesla’s 2024 annual shareholder meeting, Musk characterized himself as “pathologically optimistic,” then claimed the humanoid robots would lift the company’s market cap to $25 trillion at an unspecified future date.

In recent weeks, Musk told CNBC’s David Faber that Tesla is now training its Optimus systems to do “primitive tasks,” like picking up objects, open a door or throw a ball.

Competitors in the space include Boston Dynamics, Agility Robotics, Apptronik, 1X and Figure.

Kovac had previously served as the company’s director of Autopilot software engineering. He rose to lead the company’s Optimus unit as vice president in 2022.

Musk personally thanked Kovac for his “outstanding contributions” to the business.

Tesla didn’t respond to a request for comment.

WATCH: Tesla panel talks impact of Musk’s feud with Trump

'Closing Bell Overtime' Tesla panel talks impact of Elon Musk's feud with Pres. Trump

Technology

Tesla already had big problems. Then Musk went to battle with Trump

Published

4 hours ago

June 6, 2025

admin

Tesla already had big problems. Then Musk went to battle with Trump

President Donald Trump holds a news conference with Elon Musk to mark the end of the Tesla CEO’s tenure as a special government employee overseeing the U.S. DOGE Service on Friday May 30, 2025 in the Oval Office of the White House in Washington.

Tom Brenner | The Washington Post | Getty Images

Tesla has been facing massive challenges trying to get back on track after a disastrous first quarter. Those headwinds strengthened considerably this week.

CEO Elon Musk officially concluded his term with the Trump administration at the end of May, hitting the 130-day mark, the maximum time allowed for a “special government employee.” On his way out the door, Musk expressed sharp criticism of the Trump’s signature spending bill that’s being debated in Congress due to its expected impact on the national debt.

What started off as a policy disagreement quickly escalated into an all-out online brawl, with Musk and President Donald Trump hurling insults at one other from their respective social media platforms. After Musk called the “one, big beautiful bill” an “abomination” and rallied his followers on X to “kill the bill,” Trump said Musk had gone “CRAZY” and threatened to end government contracts and cut off subsidies for Musk’s companies. Musk responded, “Go ahead, make my day.”

The rift sent Tesla shares plummeting 14% on Thursday, wiping out roughly $152 billion in value, the most for any day in the company’s 15 year-history on the public market. While Musk is still the richest person in the world on paper, his net worth plunged by $34 billion, according to Bloomberg’s Billionaires Index.

More importantly, the spat brought about the collapse to a relationship that blended business, politics and power in a manner virtually unprecedented in U.S. history. The ramifications to Tesla, which fell out of the trillion-dollar club on Thursday, could be severe, and not just because Trump is reportedly considering selling or giving away the red Model S he purchased in March after turning the White House lawn into a Tesla showroom.

A senior White House official told NBC News on Friday that the president was “not interested” in having a call with Musk to resolve their feud.

Trump-Musk feud: Here's what's at stake for the Tesla CEO

Ire from the Trump administration could influence everything from future regulation, investigations and government support for Tesla, to decisions on tariff exemptions the company has been seeking in order to purchase Chinese-made manufacturing equipment.

Tesla shares were badly underperforming the broader market before the Musk-Trump breakup. Revenue slid 9% in the first quarter from a year earlier, with auto revenue plummeting 20%, due to the combination of increased competition from lower-cost EV makers in China and a consumer backlash to Trump’s political activities and rhetoric.

It’s certainly not what Tesla shareholders were expecting, when they sent the stock up about 30% in the days following Trump’s election victory in November. After spending close to $300 million to return Trump to the White House, Musk was poised to have a major role in the administration and be in position to push through regulatory changes in ways that benefited his companies.

Instead, his company has suffered, and Musk’s behavior is largely to blame.

One of his most divisive actions in leading the Trump administration’s Department of Government Efficiency (DOGE) was the dismantling of USAID, which previously delivered billions of dollars of food and medicine to more than 100 countries.

Beyond the U.S., Musk has endorsed Germany’s far-right extremist party AfD, and gave a gesture that many viewed as a Nazi salute at an inauguration rally.

In response, in recent months, there were numerous cases of vandalism or arson of Tesla facilities or vehicles in the U.S., as well as waves of peaceful protests at Tesla stores and service centers in North America and Europe.

Advertisements in protest of Musk have appeared in New York’s Times Square, and at bus shelters in London, urging people to boycott Tesla, some labeling the company’s EVs as “swasticars.” The Vancouver International Auto Show even removed Tesla from its exhibitors’ list fearing the company’s presence would cause safety problems.

On top all that are President Trump’s sweeping tariffs, which have led to concerns that costs will increase for parts and materials crucial for EV production. In its first-quarter earnings report in April, Tesla refrained from promising growth this year and said it will “revisit our 2025 guidance in our Q2 update.”

Board is mum

Pension funds that invest in Tesla have said the “crisis” at the company requires a leader to work a minimum of 40 hours per week to focus on solving its problems.

Public officials are echoing that sentiment, and calling on Tesla’s board to take action.

New York City Comptroller Brad Lander said on Thursday in s statement to CNBC that the “schoolyard fight” between Trump and Musk highlights how “Tesla’s weak accountability measures and poor governance threaten not only the company’s financial stability and shareholder value, but also the future of homegrown EV production.”

Brooke Lierman, comptroller of Maryland, told CNBC in an email that the company’s board “is not doing its job to ensure that there is a CEO at Tesla who is putting the company’s interests first.”

Since Musk’s name is synonymous with Tesla, the board needs to ensure that Tesla can stand on its own regardless of who’s leading the company, she added.

“Musk’s behavior continues to threaten the future of Tesla,” Lierman said. “As long as Tesla is identified with Elon Musk and he continues to be a polarizing figure, he will continue to damage the brand which is a huge part of Tesla’s value.”

Musk didn’t respond to a request for comment. CNBC also reached out for comment to board chair Robyn Denholm and directors and executives who work in government relations and in the office of the CEO. None of them responded as of the time of publication.

Elon Musk interviews on CNBC from the Tesla Headquarters in Texas.

CNBC

Tesla investors focused on business fundamentals are justified in their skepticism.

The company has failed to roll out innovative and affordable new model EVs, while Chinese competitors like BYD have flooded the market, particularly in Europe.

Analysts at Goldman Sachs on Thursday lowered their price target on Tesla mostly due to the outlook for 2025. Deliveries this quarter are tracking lower for the U.S., the analysts noted, while European sales saw a 50% year-over-year decline in April and another double-digit drop in May. China sales from those two months were down about 20% from a year earlier.

Quality is also a problem. Tesla has announced eight voluntary recalls of the Cybertruck in 15 months due to a range of issues including software bugs and sticking accelerator pedals.

Robotaxi ready?

Musk is urging investors to largely ignore the core business and look to the future, which he says is all about autonomous vehicles and humanoid robots.

But even there, Tesla is behind. In AVs the company has ceded ground to Alphabet’s Waymo, which is operating commercial robotaxi services in several U.S. markets. After a decade of missed deadlines, Musk has promised a small launch of a Tesla driverless ride-hailing service in Austin this month.

The Austin robotaxi service will operate in a geofenced area, Musk said in a recent interview with CNBC’s David Faber, and will begin with a small fleet of just 10 to 20 Model Y vehicles with Full Self-Driving (FSD) Unsupervised technology installed. If all goes well, Musk has said, Tesla will try to rapidly expand its driverless offerings to other markets like San Francisco and Los Angeles.

Watch part 1 of CNBC's interview with Tesla CEO Elon Musk

What consumers won’t be seeing anytime soon are the Cybercab and Robovan vehicles that Tesla touted at its “We, Robot” event last year to drum up customer and investor enthusiasm.

On Friday, Milan Kovac, Tesla’s vice president of Optimus robotics, announced he was leaving after joining the company in 2016. Musk thanked him for his “outstanding contribution” in a post on X.

Still, there are plenty Tesla bulls and Musk fanboys who are believers in the CEO’s vision. The stock’s 4% rebound on Friday is a sign that some saw an opportunity to buy the dip.

“I think the real story here is the investor base of Tesla literally doesn’t care about anything,” Josh Brown, CEO of Ritholtz Wealth Management and CNBC PRO contributor, told CNBC’s “Halftime Report” Friday. “This is still a nothing matters stock.”

FundStrat’s Tom Lee said the Tesla selloff was “overdone.”

Tesla’s market cap, which is dramatically inflated relative to every other U.S. car maker, is built on Musk’s vision of Tesla’s Optimus humanoid robots doing factory work and babysitting our children, while self-driving Cybercabs and Robovans make money carting around passengers.

Morgan Stanley’s Adam Jonas wrote in a note this week that, “Tesla still holds so many valuable cards that are largely apolitical,” pointing to what he sees as the company’s “AI leadership, autonomy/robotics, manufacturing, supply chain re-architecture, renewable power, [and] critical infrastructure.”

In terms of Tesla’s existing business, the most immediate impact from what’s happening in Washington D.C., is the rollback of EV credits in the current budget bill that Musk loudly opposes and that’s struggling to find sufficient support in the Senate. There’s also the matter of the tariffs and whether Tesla is able to get preferred treatment, a proposition that seems increasingly unlikely with the Musk-Trump fallout.

Matthew LaBrot, a former Tesla staff program manager, told CNBC that he’s not surprised that Musk blew up his relationship with the president. LaBrot was terminated earlier this year after sending an open letter in protest of Musk’s divisive political activity.

“I am devastated for the country and the climate, though Elon only has himself to blame,” LaBrot said in an interview. “Back a loose canon, expect stray canon fire.”

Tesla investors can’t know at the moment how much of Musk’s energy and time will now return to his lone public company, and the business responsible for the vast majority of his wealth. Even without politics, he still has SpaceX, AI startup xAI and brain tech startup Neuralink, among other businesses.

As of Thursday, Musk still had a West Wing office that hadn’t been cleaned out, two administration officials told NBC News. The space will likely be packed up in the coming days, one of the officials said.

And while his time in the Trump camp may be over, Musk has called on his followers to form a new party in the U.S.

“Is it time to create a new political party in America that actually represents the 80% in the middle?” he wrote on X on Thursday, in a post that’s now pinned at the top of his page. According to the post, 80% of 5.6 million respondents to the unofficial poll said “yes.”

Musk’s actions this week may have caused a permanent rift with the president. But one thing is clear — his company can’t get away from the White House.

WATCH: Impact of Musk’s feud with Trump

Technology

DocuSign stock tanks 18% after company cuts billings outlook

Published

12 hours ago

June 6, 2025

admin

DocuSign stock tanks 18% after company cuts billings outlook

The Docusign Inc. application for download in the Apple App Store on a smartphone arranged in Dobbs Ferry, New York, U.S., on Thursday, April 1, 2021.

Tiffany Hagler-Geard | Bloomberg | Getty Images

Shares of DocuSign tanked 18% in trading on Friday, a day after the e-signature provider reported stronger-than-expected earnings but slashed its full-year billings outlook.

Here’s how the company performed in the fiscal first quarter, compared with estimates from analysts polled by LSEG:

Earnings per share: 90 cents, adjusted, vs. 81 cents expected
Revenue: $764 million vs. $748 million expected

Billings, a closely-watched sales metric, came in at $739.6 million in the fiscal first quarter, which ended April 30. That was lower than the $746 million expected by analysts, according to StreetAccount. It also fell short of the company’s own forecast, which guided for billings between $741 million and $751 million.

For the current fiscal year, DocuSign said it expects billings of $3.28 billion to $3.34 billion, down from a range of $3.3 billion to $3.35 billion.