logo
I Spent $125 to Generate 5 AI Videos a Day With Google's Veo 3. The Sound Sets It Apart

I Spent $125 to Generate 5 AI Videos a Day With Google's Veo 3. The Sound Sets It Apart

CNETa day ago

I am just a girl who wants to be on a warm beach but most of the time, I'm trapped behind a computer screen. So, like any reporter who tests and reviews AI, I make my days more bearable by using these AI programs to create alternative-timeline versions of myself somewhere where Jimmy Buffett is playing and you can smell the salt. Here's what Google's newest AI video model, Veo 3, came up with.
My beach bonfire party dream-turned-prompt is usually my first test as I put a new AI generator through its paces. And I admit, I had pretty low expectations for Veo 3. While I did see some social media posts gawking at Veo 3's capabilities, I've seen enough slop and hallucinations to approach with skepticism. Google's AI creative products, in particular, have always felt like a bit of an afterthought to me, something the company adds on to its extensive Gemini offerings to compete with the other tech heavyweights. But this year at the company's annual I/O developer conference, Google's Imagen 4, Veo 3 and Flow all took center stage. So I dove into Veo 3.
Without spoiling anything, I walked away from Veo feeling like this was the next natural step for Google, with one feature in particular giving the company an edge that might make it a more serious contender in the AI creative space. But there are serious limits and annoyances that I hope are addressed soon. Here's how my experience went and what you need to know.
Veo 3 availability, pricing and privacy
Veo 3 is currently available for Gemini Ultra users in the US and enterprise Vertex users. In other words, you'll need to pay up to play around with the new Veo. Ultra is Gemini's newest, priciest tier at $250 per month. (It's currently half off for $125 per month for three months.) Vertex is Google's AI enterprise platform, and you'll know if you have access to it.
If you don't want to pay hundreds of dollars for access to Google's AI video tools -- and I don't blame you -- you can try out Veo 2 with Google AI's Pro plan. I found that the one-month free trial is enough time to figure out if you want to pay the $20 per month fee to continue using it. You can check out my hands-on testing with that model for more info.
Google's Gemini privacy policy says the company can collect your info to improve its technologies, which is why it recommends not sharing any confidential information with Gemini. You also agree to Google's prohibited use policy, which outlaws the creation of abusive or illegal content.
My wild ride with Veo 3
The most impressive thing about Veo 3 is its new audio generation capabilities. You don't have to tell Gemini in your prompt that you want sound; it will automatically add it. This is a first among competitors like OpenAI's Sora and Adobe's Firefly and it certainly gives Google a huge edge.
While the AI audio is a nice perk, it isn't perfect. If you're familiar with the somewhat clunky nature of AI-generated music and dialogue, you'll be able to identify it immediately. But there were times when it flowed more naturally. The clashing metal sounds and grunts in my alien fight scene were timed perfectly to their attacks, something that would've been difficult to add on my own afterward. But the dinosaur-like aliens also literally say "roar" and "hiss" instead of making those noises.
My kayaker's paddling very nearly matched up with the water sloshing sound. The nature ambience in that video was particularly lovely and added a layer of depth that's been missing from AI videos. My dream beach bonfire partiers didn't sound like any party I've ever been to, but still, points for being first and relatively unproblematic.
Of course, while the audio was nice, it doesn't take away from the weird eccentricities that continue to plague AI generators. I ran into a few hiccups, mostly with people's faces, a notoriously hard thing for AI to mimic. But compared to the glaringly obvious errors I ran into with Veo 2, the new generation does appear to have made real improvements as Google claimed it did.
I run into hallucinations a lot when I'm testing AI image and video generators, so the first thing I do is look for whether a service gives me the ability to edit it. Veo 3 doesn't offer any of these, which is a bummer. It's certainly something that's going to make it less useful for professional creators, who are used to more fine-tuning editing tools and need to make precise tweaks for their projects. You can send a follow-up prompt asking for specific changes. For example, I asked Veo to change the angle in the previous video so I could see her face, which the program handled well.
With Veo 3, you'll typically have to wait 3 to 5 minutes for a new, edited video to load, though. Veo 3 has the longest generation time of any AI video generator I've tested. But the addition of audio to the videos excuses the longer wait time in my eyes.
The worst part of Veo 3 is how quickly I hit my daily generation limit. After only five videos, I was barred for an entire 24-hour period -- something that really annoyed me and made it much harder to assess. Google's VP of Gemini and Google Labs, Josh Woodward, said in a post on X/Twitter that Ultra subscribers like me have the highest number of generations that reset daily, in the regular Gemini app and in Flow. And for me, that limit in Gemini was five videos. Flow's limit is 125, according to Woodward.
I reached out to Google to get clarity on what the daily limit is for Ultra users creating through Gemini that Woodward mentions. Here's the response: "Google AI Ultra subscribers get the highest level of access to Veo 3, our state-of-the-art video generation model, which they can use in both the Gemini app and Flow, our new AI filmmaking tool."
The limits are another sign that this isn't a tool meant for professional creation and iterative editing. You need to spend time thoughtfully crafting your prompt and if Google flubs a face or glitches, you're likely to run out of credits fast and end up out of luck. Veo 3 is better suited for AI enthusiasts who want to dip their toes in video creation, not creators experimenting with AI.
Is Veo 3 worth the cost?
After an underwhelming experience with Veo 2, I had reservations about what to expect in the usefulness and accuracy of Veo 3. But the new model was impressive, the audio especially, even though it's still missing some key features.
Let me be clear: There is no rational reason to spend hundreds of dollars on a Gemini Ultra plan only to use Veo 3. If you want to dabble for fun, you can do that with Veo 2 for hundreds less per month, and if you're a creative professional, Veo 3 still lacks crucial features like editing. The Ultra plan does offer other features, like YouTube Premium, 30 terabytes of space and access to the newest Gemini models. So if you want any of those things, then, yeah, pay up and go play around with Veo 3. But it's not worth it on its own.
Veo 3 isn't the revolutionary upgrade those social media posts might lead you to believe. It is the next generation, better than last month's Veo 2, and it shows real promise in Google's future AI video endeavors. But be prepared to pay up if you want to try it out.

Orange background

Try Our AI Features

Explore what Daily8 AI can do for you:

Comments

No comments yet...

Related Articles

Affirm Stock Down As Klarna's Buy Now, Pay Later Credit Loss Rises 17%
Affirm Stock Down As Klarna's Buy Now, Pay Later Credit Loss Rises 17%

Forbes

time25 minutes ago

  • Forbes

Affirm Stock Down As Klarna's Buy Now, Pay Later Credit Loss Rises 17%

Despite recent growth, some investors see a darker future for the industry Affirm Holdings stock is down 17% in 2025 after predicting lower-than-expected growth for the current quarter While the default rate on Buy Now, Pay Later loans rises, industry executives say they're not worried A weak first quarter gross domestic product report bodes ill for the industry Shares of Affirm Holdings — a provider of Buy Now, Pay Later loans — have lost 17% of their value in 2025, according to Google Finance. In the last three years, Affirm stock has risen considerably — rising 126% to $52 a share — and the industry has expanded faster than 50% annually. After reporting solid revenue and profit growth in the company's third quarter — while issuing a weak forecast — is the stock a bargain? Earlier this month, the stock fell 13% on Affirm's weak forecast and its bet on 0% loans, according to CNBC. The bearish case is bolstered by rising BNPL default rates at Klarna and a weaker economy which could add to the bad loans. Bulls admire the company's industry leadership and long-term approach to running the company — a view Affirm reinforces. 'It took consumers and merchants and sort of the universe about a decade to figure out what we are and just how different and important what we have found to work really is,' Affirm founder and CEO Max Levchin told CNBC. Wall Street sees the stock as significantly undervalued. Affirm shares trade 29% below $67.18 — the average price target of of 21 Wall Street analysts, TipRanks wrote. In Affirm's fiscal third quarter, revenue met expectations, profit exceeded them and its revenue forecast for the fourth quarter fell short. Here are the key numbers: Affirm's online loans rise or fall depending on the level of consumer spending on electronics, apparel and travel. The company has been aiming to add new customers — reaching 22 million in the third quarter, a 10% growth rate, noted CNBC. Through partnerships with Apple , Amazon and Shopify, GMV for The Affirm Card rose 115% from the year before while the number of active cardholders more than doubled. business is closely tied to consumer spending, as its online loan offering has become popular with sellers of electronics, apparel and travel. The company is also offering 0% interest loans in which merchants — and sometimes manufacturers — boost sales by subsidizing borrowing costs to drive sales. Such loans increased 44% — serving as an alternative to a traditional merchant discount. 'It may be an expensive net discount rate, but it's better than 10% off,' Affirm Chief Financial Officer Rob O'Hare told CNBC. Affirm says these loans extend the lifetime value of its customers. 'Every time we sign someone new through a 0% promo, some number of months or quarters from now, that's a prime candidate for the Affirm Card, and that's a lifetime value booster,' Levchin said on the company's Q3 earnings call. The growth in BNPL loans has been significant in recent years. This growth has drawn new investment into the industry and loan default rates are rising for some large participants. Since 2021, the BNPL business has accelerated at a 55% average annual rate from $97 billion, according to my June 2022 Forbes post, to $560 billion in 2025, according to Research and Markets. To finance that growth, Affirm has been securitizing — bundling and selling — some 30% of its loans. More recently, the rise of private credit has enabled Affirm to sell loans directly to institutions. These include insurers such as Liberty Mutual and Prudential. Moreover, this year private credit firm Sixth Street initiated a three year deal to buy $4 billion of Affirm's loans, according to the Wall Street Journal. Recent data suggest BNPL credit problems could rise. How so? Nearly two-thirds of BNPL loans went to borrowers with risky credit scores, according to a January report from the Consumer Financial Protection Bureau. 'Americans were using 'buy now, pay later' as a Band-Aid on top of their credit card debt,' Julie Margetta Morgan, a former CFPB official who is now president of the Century Foundation, told the Times. 'We look at it as a kind of bellwether of risks to the overall economy,' she added. BNPL providers downplay these risks. For example, Klarna — the privately held Stockholm-based BNPL provider which recently paused its IPO — suffered a 17% rise in credit losses in May. Klarna said the losses were trivial. 'There's nothing troubling or worrisome from this data,' company spokeswoman Clare Nordstrom told the New York Times. Affirm was similarly upbeat. 'We really aren't seeing anything we would label as signs of stress with our borrowers,' O'Hare said, according to the Times. BNPL customers would be especially vulnerable if the economy worsened — which is why during the Biden era, the CFPB 'called for measures to safeguard them,' the Times wrote. Unfortunately, the economy contracted in the first quarter of 2025 — with gross domestic product falling at a 0.2% rate, according to the Bureau of Economic Analysis. As U.S. household finances get worse, BNPL consumers and providers could suffer. 'Consumers are going to be squeezed and more reliant on these products,' Morgan explained to the Times, 'and the companies are being offered a free pass to construct those products in ways that are the most profitable to them.' Given the 29% upside implicit in Affirm's price target, the bulls may prevail over the bears. Affirm bears argue the company's profitability fell short of expectations because the lower growth in GMV due to a surge in 0% APR loans was not sufficient to offset their revenue less transaction costs. This is why Affirm fell short of investor expectations. The rise in 0% loans 'led to a lower take rate and RLTC margin than most forecasts,' Citizens wrote, according to CNBC. Two other analysts remain bullish on Affirm. Goldman called the company a 'strong category leader in BNPL and a share gainer vs. legacy credit providers,' noted CNBC. Barclays is bullish on recent partnerships such as the one between Affirm and Costco. Affirm sees consumers continuing to spend despite uncertainty. 'People are stressed out about the economy, yet they're shopping, they're buying, and they're paying their bills — at least they're paying their bills back to us on time,' Levchin told CNBC.

Pacers Starting Five Had Worst Group Outing In Game 5, Must Improve
Pacers Starting Five Had Worst Group Outing In Game 5, Must Improve

Forbes

time26 minutes ago

  • Forbes

Pacers Starting Five Had Worst Group Outing In Game 5, Must Improve

NEW YORK, NEW YORK - MAY 29: Pascal Siakam #43 of the Indiana Pacers drives to the basket against ... More Josh Hart #3 of the New York Knicks during the third quarter in Game Five of the Eastern Conference Finals of the 2025 NBA Playoffs at Madison Square Garden on May 29, 2025 in New York City. (Photo by) NEW YORK – With a possible berth in the NBA Finals on the table for the Indiana Pacers on Thursday night, they had an offensive performance that made even the team scratch their heads and wonder what exactly went wrong. The Pacers entered Game 5 with a 3-1 series lead. After a terrific Game 4 in which three key players stepped up for the blue and gold, their confidence was high heading back to New York City. They entered the day 2-0 in Madison Square Garden during the series and could close out the Knicks with one more win in the World's Most Famous Arena. But by the end of the night, those sanguine feelings would be gone. It was the Knicks' night. They were the much better team, and they became the first group in this playoffs to hold Indiana under 100 points in a game. The Pacers walked out of MSG after putting up just 94 points in the 48-minute battle, a bizarre performance given everything at stake. The visitors shot just 40.5% from the field, their lowest figure in the postseason by nearly three percentage points. They had a playoff-high 20 turnovers, too. Indiana's total chances were limited – and they didn't convert the ones they got. 'We obviously didn't play with the level of force that we needed to. We lost the rebound battle, we lost the turnover battle,' Pacers head coach Rick Carlisle explained postgame before even fielding a question. 'We didn't shoot well, they had a lot to do with it, so give them credit.' It was a dreadful performance all around. The Pacers trailed by double-digits from late in the second quarter onward. Any run that cut into the Knicks lead was immediately answered. Indiana is used to being the team that can respond to adversity at any moment in the postseason, but it was New York playing that way in Game 6. Several strategic reasons influenced the Pacers' poor Game 5 – the Knicks were applying more ball pressure and face guarding star ball handler Tyrese Haliburton, for example. The improvements of the Knicks, combined with a substandard night from the Pacers, led to a blowout. Most of the problems Indiana faced stemmed from the worst postseason outing for their starting five. That unit has played 241 minutes in the playoffs, and they've outscored opponents by 71 points. The five-man lineup consists of Haliburton, Andrew Nembhard, Aaron Nesmith, Pascal Siakam, and Myles Turner. They have just as many 2025 postseason outings with a plus-minus above +10 as they do with a figure below zero (six each). There is always confidence in that five-man group when they are together. They entered Game 5 a plus-eight in the Eastern Conference Finals. Yet after three quarters of play in that fifth game, they had combined to score 29 points. Knicks star Jalen Brunson had 30 points by himself. Some players were having an off night. One was injured. A few couldn't find their usual level of impact due to defensive changes from New York. For many reasons, the Pacers starting five was poor in Game 5, and their series lead dropped to 3-2 as a result. 'We weren't great as a group. I thought we lost the margin battle today,' Haliburton said over a half hour after walking off the court with a confident grin, even in defeat. 'We've got to be better as a group. I think our pace has to be better.' Indiana Pacers center Myles Turner (33) passes the ball against the New York Knicks during the ... More second quarter of Game 5 of the NBA basketball Eastern Conference final, Thursday, May 29, 2025, in New York. (AP Photo/Adam Hunger) It all starts with Haliburton. In Game 4, he had the best night of his NBA career – posting a triple-double to put his team one win away from an NBA Finals berth. Two days later, he took seven shots and finished with just eight points and six assists. Haliburton saw new coverages. Mikal Bridges was face guarding him all over the floor, making it hard for Haliburton to catch the ball on the move. Other offensive release valves had tough nights for the Pacers and couldn't punish that adjustment. It's not unusual for ramped-up pressure to slow the Pacers star, but it shouldn't have been as impactful as it ended up being. 'Rough night for me,' Haliburton stated. 'I've got to be better setting the tone, getting downhill. I feel like I didn't do a great job of that.' Haliburton's personal effectiveness has dropped in the past when he sees this type of coverage. In this instance, it also limited the ability of his teammates. Their star point guard had fewer avenues to get into the paint and distribute, so easy shots were harder to come by. Play finishers were hurting as a result. Nesmith, who started the series on fire and is dealing with an ankle injury suffered in Game 3, made just one of his eight field goal attempts. Turner only took three shots. Nembhard, who can be more than just a finisher but didn't find much space while being defended by OG Anunoby, went 3/8 from the field. Those three all struggled and combined to score 14 points on 19 shot attempts. Nesmith gets more of a pass because of his injury. 'It's not 100%, but it's no excuse. I've got to be better,' he said of his sprained ankle. But he was still terrific in Game 4. Being off from the field in Game 5 and only playing for 15:52 challenged the usually-effective starting five. The Pacers will hope he's better in any future battles. 'It still comes down to the physicality from the jump, just playing with more effort and energy and worrying about us more than them,' Nesmith said matter-of-factly at his locker postgame. Turner wasn't himself either. He took eight or more shots in the first four games of the series but only found room for three in Game 5. His shots usually come from strong ball movement and offensive flow, so that field goal attempt number being so low shows just how disconnected the starting five was Thursday night. Turner's impact has been muted in the last few games and needs to pop. Siakam can get the Pacers going on some nights when their offense is subpar. That's how the blue and gold won Game 2 of this series. But even he had an off night, shooting just 5/13 and missing four attempts from the foul line. 'They were more aggressive from the jump. They brought the fight to us. I just didn't think we brought it enough,' Siakam said. Altogether, it was the starting lineup's worst game this series. They played for under 10 minutes together for the first time this postseason – though Nesmith's injury was a factor in that. It was only the second time in these playoffs that they were outscored by 10+ points while sharing the floor. The bench was good enough for Indiana in Game 5. The starters were not. They shot 33.3% and had 10 turnovers. If the Pacers adjust like they have all season, they'll be fine. The team hasn't lost consecutive games since early March. But after the worst collective performance from their starting five in the postseason, the blue and gold need more from their best players to close out the Knicks.

Google quietly released an app that lets you download and run AI models locally
Google quietly released an app that lets you download and run AI models locally

Yahoo

time26 minutes ago

  • Yahoo

Google quietly released an app that lets you download and run AI models locally

Last week, Google quietly released an app that lets users run a range of openly available AI models from the AI dev platform Hugging Face on their phones. Called Google AI Edge Gallery, the app is available for Android and will soon come to iOS. It allows users to find, download, and run compatible models that generate images, answer questions, write and edit code, and more. The models run offline, without needing an internet connection, tapping into supported phones' processors. AI models running in the cloud are often more powerful than their local counterparts, but they also have their downsides. Some users might be wary of sending personal or sensitive data to a remote data center, or want to have models available without needing to find a Wi-Fi or cellular connection. Google AI Edge Gallery, which Google is calling an "experimental Alpha release," can be downloaded from GitHub by following these instructions. The home screen shows shortcuts to AI tasks and capabilities like "Ask Image" and "AI Chat." Tapping on a capability pulls up a list of models suited for the task, such as Google's Gemma 3n. Google AI Edge Gallery also provides a "Prompt Lab" users can use to kick off "single-turn" tasks powered by models, like summarizing and rewriting text. The Prompt Lab comes with several task templates and configurable settings to fine-tune the models' behaviors. Your mileage may vary in terms of performance, Google warns. Modern devices with more powerful hardware will predictably run models faster, but the model size also matters. Larger models will take more time to complete a task — say, answering a question about an image — than smaller models. Google's inviting members of the developer community to give feedback on the Google AI Edge Gallery experience. The app is under an Apache 2.0 license, meaning it can be used in most contexts — commercial or otherwise — without restriction. 登入存取你的投資組合

DOWNLOAD THE APP

Get Started Now: Download the App

Ready to dive into the world of global news and events? Download our app today from your preferred app store and start exploring.
app-storeplay-store