โŒ

Normal view

Received before yesterday

Where AI meets design: Runway co-founder Alejandro Matamala Ortiz takes the AI Stage at TechCrunch Disrupt 2025

10 July 2025 at 22:45
TechCrunch Disrupt 2025 is the epicenter where 10,000+ startup and VC leaders gather to explore the future of innovation โ€” and this year, two AI Stages bring the intersection of design and machine learning into sharp focus. Weโ€™re thrilled to welcome Alejandro Matamala Ortiz, co-founder and chief design officer at Runway, to one of the [โ€ฆ]

How to actually raise a seed round: Actionable advice from top investors at TechCrunch Disrupt 2025

10 July 2025 at 22:15
This isnโ€™t your typical venture panel. Expect sharp insights on how to cut through the noise, structure your raise, and build relationships that last beyond a term sheet. Whether you're gearing up for your first round or advising a founder who is, this is the TC Disrupt session you donโ€™t want to miss.

Belkin ends support for most Wemo devices and its Wemo app

10 July 2025 at 21:13
Another day, another smart device bites the dust -- or, in this case, a full lineup of smart devices. On Thursday, Belkin announced it will end technical support for its older Wemo products as of January 31, 2026, and the app used to control the devices will no longer be supported. The decision will impact a range of devices, including smart plugs, light switches, smart bulbs, baby monitors, kitchen appliances, heaters, air purifiers, motion sensors, and more.

5 days until TechCrunch All Stage โ€” save up to $475 before prices rise

10 July 2025 at 18:00
In just five days, startup leaders from across the country will descend on Bostonโ€™s SoWa Power Station for TechCrunch All Stage 2025 โ€” and your chance to lock in the lowest ticket prices will be gone.

US appeals court blocks FTCโ€™s โ€˜click-to-cancelโ€™ rule for subscriptions

10 July 2025 at 14:01
A U.S. appeals court has blocked the Federal Trade Commission's โ€œclick-to-cancelโ€ rule that would have required companies to make it as easy to cancel a subscription as it was to sign up.

How exactly did Grok go full 'MechaHitler?'

10 July 2025 at 15:10

Earlier this week, Grok, X's built-in chatbot, took a hard turn toward antisemitism following a recent update. Amid unprompted, hateful rhetoric against Jews, it even began referring to itself as MechaHitler, a reference to 1992's Wolfenstein 3D. X has been working to delete the chatbot's offensive posts. But it's safe to say many are left wondering how this sort of thing can even happen.

I spoke to Solomon Messing, a research professor at New York University's Center for Social Media and Politics, to get a sense of what may have gone wrong with Grok. Before his current stint in academia, Messing worked in the tech industry, including at Twitter, where he founded the company's data science research team. He was also there for Elon Musk's takeover.

The first thing to understand about how chatbots like Grok work is that they're built on large language models (LLMs) designed to mimic natural language. LLMs are pretrained on giant swaths of text, including books, academic papers and, yes, even social media posts. The training process allows AI models to generate coherent text through a predictive algorithm. However, those predictive capabilities are only as good as the numerical values or "weights" that an AI algorithm learns to assign to the signals it's later asked to interpret. Through a process known as post-training, AI researchers can fine-tune the weights their models assign to input data, thereby changing the outputs they generate.

"If a model has seen content like this during pretraining, there's the potential for the model to mimic the style and substance of the worst offenders on the internet," said Messing.

In short, the pre-training data is where everything starts. If an AI model hasnโ€™t seen hateful, anti-antisemitic content, it wonโ€™t be aware of the sorts of patterns that inform that kind of speech โ€” including phrases such as "Heil Hitler" โ€” and, as a result, it probably won't regurgitate them to the user.

In the statement X shared after the episode, the company admitted there were areas where Grok's training could be improved. "We are aware of recent posts made by Grok and are actively working to remove the inappropriate posts. Since being made aware of the content, xAI has taken action to ban hate speech before Grok posts on X," the company said. "xAI is training only truth-seeking and thanks to the millions of users on X, we are able to quickly identify and update the model where training could be improved."

Elon Musk said users would
Screenshots via X

As I saw people post screenshots of Grok's responses, one thought I had was that what we were watching was a reflection of X's changing userbase. It's no secret xAI has been using data from X to train Grok; easier access to the platform's trove of information is part of the reason Musk said he was merging the two companies in March. What's more, X's userbase has become more right wing under Musk's ownership of the site. In effect, there may have been a poisoning of the well that is Grok's training data. Messing isn't so certain.

"Could the pre-training data for Grok be getting more hateful over time? Sure, if you remove content moderation over time, the userbase might get more and more oriented toward people who are tolerant of hateful speech [...] thus the pre-training data drifts in a more hateful direction," Messing said. "But without knowing what's in the training data, it's hard to say for sure."

It also wouldn't explain how Grok became so antisemitic after just a single update. On social media, there has been speculation that a rogue system prompt may explain what happened. System prompts are a set of instructions AI model developers give to their chatbots before the start of a conversation. They give the model a set of guidelines to adhere to, and define the tools it can turn to for help in answering a prompt.

In May xAI blamed "an unauthorized modification" to Grok's prompt on X for the chatbot's brief obsession with "white genocide" in South Africa. The fact that the change was made at 3:15AM PT made many suspect Elon Musk had done the tweak himself. Following the incident, xAI open sourced Grok's system prompts, allowing people to view them publicly on GitHub. After Tuesday's episode, people noticed xAI had deleted a recently added system prompt that told Grok its responses should "not shy away from making claims which are politically incorrect, as long as they are well substantiated."

Messing also doesn't believe the deleted system prompt is the smoking gun some online believe it to be.

"If I were trying to ensure a model didn't respond in hateful/racist ways I would try to do that during post-training, not as a simple system prompt. Or at the very least, I would have a hate speech detection model running that would censor or provide negative feedback to model generations that were clearly hateful," he said. "So it's hard to say for sure, but if that one system prompt was all that was keeping xAI from going off the rails with Nazi rhetoric, well that would be like attaching the wings to a plane with duct tape."

He added: "I would definitely say a shift in training, like a new training approach or having a different pre-training or post-training setup would more likely explain this than a system prompt, particularly when that system prompt doesnโ€™t explicitly say, 'Do not say things that Nazis would say.'"

On Wednesday, Musk suggested Grok was effectively baited into being hateful. "Grok was too compliant to user prompts," he said. "Too eager to please and be manipulated, essentially. That is being addressed." According to Messing, there is some validity to that argument, but it doesn't provide the full picture. "Musk isnโ€™t necessarily wrong," he said, "Thereโ€™s a whole art to 'jailbreaking' an LLM, and itโ€™s tough to fully guard against in post-training. But I donโ€™t think that fully explains the set of instances of pro-Nazi text generations from Grok that we saw."

If there's one takeaway from this episode, it's that one of the issues with foundational AI models is just how little we know about their inner workings. As Messing points out, even with Meta's open-weight Llama models, we don't really know what ingredients are going into the mix. "And that's one of the fundamental problems when we're trying to understand what's happening in any foundational model," he said, "we don't know what the pre-training data is."

In the specific case of Grok, we don't have enough information right now to know for sure what went wrong. It could have been a single trigger like an errant system prompt, or, more likely, a confluence of factors that includes the system's training data. However, Messing suspects we may see another incident just like it in the future.

"[AI models] are not the easiest things to control and align," he said. "And if you're moving fast and not putting in the proper guardrails, then you're privileging progress over a sort of care. Then, you know, things like this are not surprising."

This article originally appeared on Engadget at https://www.engadget.com/ai/how-exactly-did-grok-go-full-mechahitler-151020144.html?src=rss

ยฉ

ยฉ Igor Bonifacic for Engadget

A closeup of the Grok icon on iOS.

Google's Gemini app can now generate videos from static images

10 July 2025 at 15:00

Starting today, Google is bringing image-to-video generation to the Gemini app. The feature comes courtesy of the company's Veo 3 model, which Google began rolling out more broadly to AI Pro users last week after it was initially only available to AI Ultra subscribers.

To start using Gemini's image-to-video generation, click the "tools" option in the prompt bar and then select "video." Google is currently limiting Veo 3 to producing eight-second clips at 720p. Gemini will output your request in a 16:9 landscape format, so the resulting clips won't be great for sharing on social media โ€” unlike those generated by TikTok's AI Alive feature, for example. However, Veo 3 is currently one of the only AI models capable of generating synced audio alongside the video it creates.

You can also use Veo 3's image-to-video generation feature in Flow, Google's AI filmmaking app. As of today, the program is available in 75 additional countries. Over in the Gemini app, image-to-video generation is rolling out on the web today. Google expects most mobile users will have access by the end of the week. A $20 per month Google AI Pro or $250 per month AI Ultra subscription is required to use the new feature.

This article originally appeared on Engadget at https://www.engadget.com/ai/googles-gemini-app-can-now-generate-videos-from-static-images-150052396.html?src=rss

ยฉ

ยฉ Google

Google's Veo 3 model transforms an image of a cardboard box into a video where that box erupts with colorful confetti.

I was scared to leave NYC — but I moved to Nashville, fell in love, and have been happily living here for a decade

10 July 2025 at 14:39
Woman with baby in carrier on her chest waiting for subway
It wasn't easy to leave New York City, but moving to Nashville has been a huge, great step in my life.

Amelia Edelman

  • I thought I'd live in New York City forever, but I hit my breaking point and moved to Nashville.
  • I could enjoy many things I love about city life and get more space for less. I even fell in love.
  • It's been about a decade since my move, and I'm happily living here with my husband and two kids.

New York or nowhere. It's a T-shirt and an Instagram, but it was also my personal motto for most of my young life.

I was born in the Bronx, got my first post-college apartment in Queens, spent nearly a decade in a fifth-floor walk-up in Manhattan, and brought my first baby home to Brooklyn.

In high school and college, I spent time living in Connecticut, Poughkeepsie, and Scotland, but always felt the draw back to NYC.

By age 30, I'd spent most of my life in the city, and was living my own NYC dream working at a buzzy women's media company.

I had never imagined living anywhere else. Then, I hit my breaking point.

After a reality check, I gave myself permission to leave New York

Woman sleeping on bus with baby on her lap
Being a single mom in New York City came with challenges.

Amelia Edelman

New York wasn't just my city; it was a huge part of my identity.

However, I was burned out at my job, underpaid, and commuting hours on the subway between Manhattan and my shoebox of an apartment in Crown Heights.

I was paying a nanny most of my salary just so I could have the privilege of โ€ฆ not seeing my newborn.

After each day speed-editing dozens of articles and pumping breastmilk in a closet at the office, I would sprint to the subway at 7 p.m. in hopes of seeing my son while he was still awake.

I would never make it back in time. I'd kiss his sleeping face, pay the nanny, and cry.

By the time my son outgrew his bassinet and needed to transition to a crib, it became clear my tiny apartment was too small for us.

A crib and an adult bed didn't fit in the space, so I gave the latter away and spent the last six months of my New York life sleeping on a bedroll on the floor.

And I finally gave myself permission to consider the impossible: leaving. I just wasn't sure where to go next.

Nashville wasn't the plan, but it was the answer

Downtown Nashville skyline along water
Nashville seemed like a city I could really enjoy living in.

RudyBalasko/Getty Images

I knew I wanted to live in a city, but I needed somewhere cheaper (and way more chill) than New York.

I didn't want to relive my teen years in the Connecticut suburbs, or even that blissful but too-quiet year in college when I lived on the coast of northern Scotland.

I wanted my son to grow up in a real community: walking to public school and the playground and pizza parlor like I did as a little kid in the Bronx. I wanted to take him to museums and music venues.

Soon, Nashville was on my radar โ€” once I factored in my other wants, it seemed like the biggest, most diverse, most affordable city I could afford.

I told my employer I was moving, and that I could quit or they could let me go remote. They let me keep my job. I bought a four-bedroom house in East Nashville with a monthly mortgage that was close to half my rent in Brooklyn.

My new block had coffee shops, bars, a pharmacy, a pizza parlor, a bodega, and a vintage store that was also an art gallery that was also a music venue. So Brooklyn! I felt right at home.

Kid walking down empty street in Nashville during sunset
My life moves at a slower pace in Nashville than it did in New York City, but I've gotten used to it.

Amelia Edelman

Sure, at first everything felt โ€ฆ slow. I didn't live near downtown, so the bustle dial was turned way down.

Initially, it was hard to sleep without sirens and shouting outside my window. But as the weeks turned into months, I started to notice I was breathing easier.

Nashville gave me more space โ€” not just physical space (for a crib and a bed, imagine!) but space in my day that was no longer spent commuting, hauling a stroller up and down stairs, and rushing to the laundromat.

It gave me more accessible green spaces than New York had; my son and I could be out on a hike within 20 minutes, no Metro-North train ride necessary.

Without a long commute, I had time to make real dinners, to lounge on porches, and to get to know my neighbors. I made friends, joined a nonprofit, and started teaching yoga at the local studio.

I had the emotional space to date around casually and have fun.

When my son was 2 ยฝ, I met one particular musician. He was calm but passionate, goofy but grounded, Southern polite but also punk rock. He loved my son.

By year five in Nashville, we were married. Year six, he adopted my son. That same year, our second son was born.

Moving was the best decision I was scared to make

House with snow on its roof, lawn, and a kid out front
I've enjoyed raising my kids in Nashville.

Amelia Edelman

There's a common fear among people who leave big cities that we're somehow giving up. I definitely felt it.

I worried that moving to a smaller city would mean trading ambition for comfort. My work changed, yes.

I later shifted away from a traditional media job into freelance and consulting work, but I'm making more money now since I'm paid per project rather than being expected to work endless hours for an unchanging salary.

Now, I work smarter, not harder. I live smarter. I've stopped defining myself solely by my ever-climbing corporate media job title, or my precious 917 area code.

Nashville gave me the space to grow in unexpected directions. I have a garden, I volunteer, and I made friends who didn't care about who I worked for. I built a community that is unparalleled in its supportive and radically inclusive nature.

This city isn't perfect, but it's become home. At the time, leaving New York felt like the biggest risk of my life. Today, I think of how scared I was of the best decision I ever made, and laugh.

It's been nearly a decade since I left New York, and although I still visit my "hometown" often and miss it dearly sometimes, I don't regret the move for a second.

Well, maybe I just regret not leaving 10 years earlier.

Read the original article on Business Insider

AI models aren't made equal. Some nonprofits are creating their own tools instead.

10 July 2025 at 13:31
Education Above All Foundation

Courtesy of Education Above All Foundation; Alyssa Powell/ BI

  • Nonprofits like Education Above All are using AI to address global inequities.
  • AI initiatives align with the UN Sustainable Development Goals to promote peace and prosperity.
  • This article is part of "How AI Is Changing Everything," a series on AI adoption across industries.

As millions of young people worldwide increasingly rely on AI chatbots to acquire knowledge as part of their learning โ€” and even complete assignments for them โ€” one organization is concerned that those in developing countries without access to the tech could be put at an unfair disadvantage.

And it's using the very technology it believes is causing this problem to fix it.

Education Above All, a nonprofit based in Qatar, believes that because most of the world's popular AI chatbots are created in Silicon Valley, they aren't equipped to understand the linguistic and ethnic nuances of non-English-speaking countries, creating education inequities on a global scale. But its team sees AI as a way to tackle this problem.

In January 2025, the charity teamed up with MIT, Harvard, and the United Nations Development Programme to introduce a free and open-source AI literacy program called Digi-Wise. Delivered in partnership with educators in the developing world, it encourages children to spot AI-fueled misinformation, use AI tools responsibly in the classroom, and even develop their own AI tools from scratch.

As part of this, the charity has developed its own generative AI chatbot called Ferby. It allows users to access and personalize educational resources from the Internet-Free Education Resource Bank, an online library containing hundreds of free and open-source learning materials.

Education Above All said it's already being used by over 5 million Indian children to access "project-based learning" in partnership with Indian nonprofit Mantra4Change. More recently, Education Above All has embedded Ferby into edtech platform SwiftChat, which is used by 124 million students and teachers across India.

"Ferby curates, customizes, and creates learning materials to fit local realities, so a teacher in rural Malawi can run the right science experiment as easily as a teacher in downtown Doha," said Aishwarya Shetty, an education specialist at Education Above All. "By marrying offline ingenuity with AI convenience, we make learning local, low-resource, and always within reach, yet at scale."

Education Above All is among a group of organizations using AI to tackle global inequality and work toward realizing the United Nations Sustainable Development Goals. Created in 2015, the UN SDGs comprise 17 social, economic, and environmental targets that serve as guidelines for nations, businesses, and individuals to follow to help achieve a more peaceful and prosperous world. Education Above All's projects fall under SDG 4: inclusive and equitable education.

A global effort

A range of other organizations are using AI to augment and enhance their education programming.

Tech To The Rescue, a global nonprofit that connects charities with pro-bono software development teams to meet their goals, is another organization using AI in support of the UN SDGs. Last year, it launched a three-year AI-for-good accelerator program to help NGOs meet the various UN SDGs using AI.

One organization to benefit from the program is Mercy Corps, a humanitarian group that works across over 40 countries to tackle crises like poverty, the climate crisis, natural disasters, and violence. Through the accelerator, it created an AI strategy tool that helps first responders predict disasters and coordinate resources. The World Institute on Disability AI also participated in the accelerator program, creating a resource-matching system that helps organizations allocate support to people with disabilities in hours rather than weeks.

Similarly, the International Telecommunication Union โ€” the United Nations' digital technology agency, and one of its oldest arms โ€” is supporting organizations using technology to achieve the UN SDGs through its AI for Good Innovation Factory startup competition. For example, an Indian applicant โ€” a startup called Bioniks โ€” has enabled a teenager to reclaim the ability to do simple tasks like writing and getting dressed through the use of AI-powered prosthetics.

Challenges to consider

While AI may prove to be a powerful tool for achieving the UN SDGs, it comes with notable risks. Again, as AI models are largely developed by American tech giants in an industry already constrained by gender and racial inequality, unconscious bias is a major flaw of AI systems.

To address this, Shetty said layered prompts for non-English users, human review of underlying AI datasets, and the creation of indigenous chatbots are paramount to achieving Education Above All's goals.

AI models are also power-intensive, making them largely inaccessible to the populations of developing countries. That's why Shetty urges AI companies to provide their solutions via less tech-heavy methods, like SMS, and to offer offline features so users can still access AI resources when their internet connections drop. Open-source, free-of-charge subscriptions can help, too, she added.

AI as a source for good

Challenges aside, Shetty is confident that AI can be a force for good over the next few years, particularly around education. She told BI, "We are truly energized by how the global education community is leveraging AI in education: WhatsApp-based math tutors reaching off-grid learners; algorithms that optimize teacher deployment in shortage areas; personalized content engines that democratize education; chatbots that offer psychosocial support in crisis zones and more."

But Shetty is clear that AI should augment, rather than displace, human educators. And she said the technology should only be used if it can solve challenges faced by humans and add genuine value.

"Simply put," she said, "let machines handle the scale, let humans handle the soul, with or without AI tools."

Read the original article on Business Insider

ChatGPT made up a product feature out of thin air, so this company created it

9 July 2025 at 21:59

On Monday, sheet music platform Soundslice says it developed a new feature after discovering that ChatGPT was incorrectly telling users the service could import ASCII tablatureโ€”a text-based guitar notation format the company had never supported. The incident reportedly marks what might be the first case of a business building functionality in direct response to an AI model's confabulation.

Typically, Soundslice digitizes sheet music from photos or PDFs and syncs the notation with audio or video recordings, allowing musicians to see the music scroll by as they hear it played. The platform also includes tools for slowing down playback and practicing difficult passages.

Adrian Holovaty, co-founder of Soundslice, wrote in a blog post that the recent feature development process began as a complete mystery. A few months ago, Holovaty began noticing unusualย activity in the company's error logs. Instead of typical sheet music uploads, users were submitting screenshots of ChatGPT conversations containing ASCII tablatureโ€”simple text representations of guitar music that look like strings with numbers indicating fret positions.

Read full article

Comments

ยฉ Malte Mueller via Getty Images

โ€œThings weโ€™ll never knowโ€ science fair highlights USโ€™s canceled research

9 July 2025 at 19:55

Washington, DCโ€”From a distance, the gathering looked like a standard poster session at an academic conference, with researchers standing next to large displays of the work they were doing. Except in this case, it was taking place in the Rayburn House Office Building on Capitol Hill, and the researchers were describing work that they werenโ€™t doing. Called "The things weโ€™ll never know," the event was meant to highlight the work of researchers whose grants had been canceled by the Trump administration.

A lot of court cases have been dealing with these cancellations as a group, highlighting the lack of scientificโ€”or seemingly rationalโ€”input into the decisions to cut funding for entire categories of research. Here, there was a much tighter focus on the individual pieces of research that had become casualties in that larger fight.

Seeing even a small sampling of the individual grants that have been terminated provides a much better perspective on the sort of damage that is being done to the US public by these cuts and the utter mindlessness of the process that's causing that damage.

Read full article

Comments

ยฉ John Timmer

AI mania pushes Nvidia to record $4 trillion valuation

9 July 2025 at 18:35

On Wednesday, Nvidia became the first company in history to reach $4 trillion market valuation as shares rose more than 2 percent, reports CNBC. The GPU maker's stock has climbed 22 percent since the start of 2025, continuing a trend driven by demand for AI hardware following ChatGPT's late 2022 launch.

The milestone marks the highest market cap ever recorded for a publicly traded company, surpassing Apple's previous record of $3.8 trillion set in December. Nvidia first crossed $2 trillion in February 2024 and reached $3 trillion just four months later in June. The $4 trillion valuation represents a market capitalization larger than the GDP of most countries.

As we explained in 2023, Nvidia's continued success has been intimately tied to growth in demand for hardware that runs AI models as capably and efficiently as possible. The company's data center GPUs excel at performing billions of matrix multiplications necessary to train and run neural networks due to their parallel architectureโ€”hardware architectures that originated as video game graphics accelerators now power the generative AI boom.

Read full article

Comments

ยฉ Nvidia / Benj Edwards

The 4 Most Dangerous Words for Bitcoin Investors: "This Time It's Different"

Key Points

  • Bitcoin historically experiences four-year boom-and-bust cycles marked by tremendous volatility.

  • According to some investors, Bitcoin has outgrown these boom-and-bust cycles.

  • While some elements of the current cycle are different this time around, Bitcoin is not immune from experiencing a significant market correction.

Famed investor John Templeton once remarked, "The four most dangerous words in investing are: This time it's different." It's an acknowledgment that investing tends to follow certain timeless rules, and that investors often become too bullish about the prospects of any high-flying asset.

That's what has me so concerned about Bitcoin (CRYPTO: BTC) right now. There's a growing sense that Bitcoin has somehow outgrown its historical boom-and-bust cycle, and that severe market corrections are now a relic of the past. But is that really the case?

Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now. Learn More ยป

The four-year Bitcoin cycle

Long-time Bitcoin investors recognize that the world's most popular cryptocurrency tends to follow four-year cycles. Each cycle is further subdivided into four distinct phases. There's a period of accumulation, followed by a period of rapid growth, a bubble, and then a crash.

The timing of this cycle might seem arbitrary, except for the fact that Bitcoin has a halving event every four years. This halving, which cuts in half the rate of new Bitcoin creation, is what leads to the period of rapid growth. If you know the date of the halving, you can basically add 12 to 18 months, and that's when the "bubble" phase should be coming to an end.

The last Bitcoin halving took place on April 19, 2024. If history is any guide, the current period of Bitcoin growth should be coming to an end sometime around October of this year. If we're lucky, the price of Bitcoin could hold its ground until January 2026.

You can see where I'm going with this -- we're "due" for a market correction. If Bitcoin follows its historical pattern, then the four-year cycle should be wrapping up relatively soon. And that means a significant market correction (or something worse) could be incoming.

"This time it's different"

The typical response to this, of course, is: "This time it's different." This is the first cycle, for example, when the spot Bitcoin exchange-traded funds (ETFs) have existed. This is the first cycle when a Strategic Bitcoin Reserve has existed. This is the first cycle when the Bitcoin treasury company model has been in vogue. This is the first cycle when the U.S. government is supporting pro-Bitcoin policies. This is the first cycle when risk-averse institutional investors such as pension funds are buying Bitcoin.

A bear wearing pink sunglasses.

Image source: Getty Images.

You could probably list several more factors that are "different" this time around. But think back to what Templeton said decades ago. There are certain timeless truths about the financial markets that you simply can't ignore. And one of those is that financial assets can't defy gravity for long. What goes up must come down.

The Bitcoin bulls will reject this line of thinking, of course. Michael Saylor, the founder and executive chairman of MicroStrategy (NASDAQ: MSTR), recently told Bloomberg TV in an interview: "Crypto winter is not coming back. We're past that phase." Saylor discounts the chances of any significant pullback, and is confident that Bitcoin will continue to soar in value. Saylor's new price target for Bitcoin is a head-spinning $21 million per coin.

Ignore the history of Bitcoin at your own peril

For any investors new to Bitcoin, I would highly recommend familiarizing yourself with the four-year Bitcoin cycle. Take a good, hard look at the long-term chart of Bitcoin. You'll see that Bitcoin has had several drawdowns of 77% or higher throughout its history. Moreover, even bullish periods of the cycle have experienced significant market corrections.

In short, the price of Bitcoin has never gone straight up. During the previous crypto bull market cycle, the price of Bitcoin hit a (then) all-time high of $69,000 in November 2021. Twelve months later, the price of Bitcoin was $16,000, a decline of 77%. The price of Bitcoin later recovered, but some crypto investors were wiped out.

Admittedly, the volatility of Bitcoin is lower these days, and the pullbacks appear to be less extreme. Even when Bitcoin fell to $75,000 earlier this year, it quickly recovered. Over the long haul, I have no doubt that Bitcoin will continue to march higher. Maybe it will one day hit the mythical $1 million mark.

However, I'm still not convinced that Wall Street has figured out Bitcoin, and that it's up only from here on out. If you buy into the logic of "this time it's different," you may be setting yourself up for disappointment later.

Should you invest $1,000 in Bitcoin right now?

Before you buy stock in Bitcoin, consider this:

The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy nowโ€ฆ and Bitcoin wasnโ€™t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.

Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, youโ€™d have $687,764!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, youโ€™d have $980,723!*

Now, itโ€™s worth noting Stock Advisorโ€™s total average return is 1,048% โ€” a market-crushing outperformance compared to 179% for the S&P 500. Donโ€™t miss out on the latest top 10 list, available when you join Stock Advisor.

See the 10 stocks ยป

*Stock Advisor returns as of July 7, 2025

Dominic Basulto has positions in Bitcoin. The Motley Fool has positions in and recommends Bitcoin. The Motley Fool has a disclosure policy.

Where Will Costco Stock Be in 3 Years?

Key Points

  • Costco stock has regularly outperformed the S&P 500 over three-year stretches.

  • The companyโ€™s membership model and low prices help it thrive in all kinds of markets.

  • The Kirkland store brand is another major driver of both value and loyalty.

Warehouse retailer Costco Wholesale (NASDAQ: COST) has been a fairly cyclical stock in the past. The stock chart often stays close to the S&P 500 (SNPINDEX: ^GSPC) index for a few years, followed by a couple of years with market-crushing returns. Taken together, this pattern has resulted in wealth-building shareholder returns over the long haul.

Where is Costco positioned in this cycle right now, and where should the stock go in the next three years? Let's take a look.

Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now. Learn More ยป

Costco keeps running circles around the S&P 500

First, let me state the obvious. Nobody knows exactly what the next three years will look like. There could be another pandemic around the next corner, or perhaps a new technology that makes the artificial intelligence and quantum computing booms look small. Historians can only look backward, and unpredictable events can unravel any forward-looking prediction I make -- bullish or bearish.

That being said, Costco has a long history of outperforming the market in a wide variety of economic situations. I compared the stock's total return to the S&P 500 (SNPINDEX: ^GSPC) index over a handful of three-year periods.

COST Total Return Level Chart

COST Total Return Level data by YCharts

Starting with a 109% 3-year Costco return against a 67% gain in the market index as of July 7, 2025, all of my sample comparisons made Costco look good.

That includes the inflation-burdened span from the end of 2022 to Dec 31, 2024. The COVID-19 crisis, starting from January 2020, showed a 63% Costco gain in three years while the S&P 500 rose 25%.

There was a close call from 2015 to 2018, with Costco lagging behind the market for most of that period -- but the warehouse retailer still came back to a 35% vs. 20% victory in a period characterized by the Brexit and a slower pace of Chinese economic growth.

I kept going with a few more samples, but the results were incredibly consistent. I'm sure there are counterexamples out there, especially if you look at spans based on mid-year dates instead of full calendar years, but the pattern is clear. Costco rarely underperforms the S&P 500 in any given three-year period.

A staff member of a warehouse store, viewing the shelves through a VR headset.

Image source: Getty Images.

How Costco's secret sauce keeps working

Past performance should never be seen as a guarantee of upcoming results. Always in motion, the future is. Even so, Costco keeps proving that the company can deliver strong results in different markets.

There are many reasons for Costco's resilience. For example:

  • Low prices make Costco a popular shopping destination when budgets are tight.
  • The membership model lets Costco sell goods with very low gross profit margins, leaving the heavy bottom-line lifting to those annual fees. That revenue stream is pretty much pure profit, with very predictable annual and quarterly volumes.
  • The Kirkland store brand is more than just another cost-cutting supply chain adjustment. Costco puts in a lot of work to ensure that the Kirkland products are competitive with the best name-brand alternatives, and you won't see this brand in sections where the research and development team can't check all the right boxes. For example, co-founder Jim Sinegal turned down the idea of selling Costco-branded gas 8 times over several years. Now, the gas pumps account for a substantial portion of Costco's total sales.

That's just a handful of stabilizing qualities. Overall, Costco is an incredibly well-run business that can roll with lots of different punches. And that's why the stock keeps outperforming even the high-quality names of the S&P 500.

Why betting against Costco is usually a bad idea

Based on Costco's history of stellar and predictable returns, I'm confident that the stock should keep up with the market over the next three years as well. The business model is both flexible and robust -- a rare combination that spells success under most circumstances.

On the downside, Costco's top-notch business is no secret. You've seen the stock chart already, and shares are changing hands at the lofty valuation of 56 times earnings. I keep kicking myself for not picking up a few Costco shares years ago, when they were more affordable. Again, the stock price more than doubled since July 2022.

Chances are, there'll be more kicking to do in 2028 if I don't grab some Costco stock this summer. Truly great companies can be worth a premium price tag, and Costco belongs in this category.

Should you invest $1,000 in Costco Wholesale right now?

Before you buy stock in Costco Wholesale, consider this:

The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy nowโ€ฆ and Costco Wholesale wasnโ€™t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.

Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, youโ€™d have $687,764!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, youโ€™d have $980,723!*

Now, itโ€™s worth noting Stock Advisorโ€™s total average return is 1,048% โ€” a market-crushing outperformance compared to 179% for the S&P 500. Donโ€™t miss out on the latest top 10 list, available when you join Stock Advisor.

See the 10 stocks ยป

*Stock Advisor returns as of July 7, 2025

Anders Bylund has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Costco Wholesale. The Motley Fool has a disclosure policy.

โŒ