In the world of entrepreneurship, one business owner’s journey proves the power of simple technology.

And group chats.

When Richelle Samy founded Culture of Stamina, a coaching service, she set out to create an online presence that was elegant and professional. She found what she was looking for when a group chat led her to Solo, Mozilla’s AI website creator for solopreneurs. After a few clicks using Solo’s generative AI (GenAI) tools, Richelle had a website for her brand with bold, sharp text and colors that perfectly captured her vision.

For Richelle, Solo enables her to focus more on empowering and training her clients instead of spending hours on her website. Other website builders weren’t as easy.

“Those tools are really nice, but I feel like you need a little bit of knowledge of what you want to do and how you want to put things together,” she recalled. “Whereas with Solo, I knew I was looking for a window for my business for people to contact me, and I only wanted a couple of pages. It was very easy to use something that was already pre-made, versus something I had to do from scratch.”

When Mozilla launched Solo in December, we were curious to see how people like Richelle would receive GenAI with website creation. Six months into this journey, we’re happy with the progress, and it’s time to reflect on what we’ve accomplished and learned along the way. We talked with the head of Solo at Mozilla, Raj Singh, about the AI website creator, its journey since the early stages, how it’s disrupting itself in the $2.1 billion website builder software industry with free custom domains and much more. Below is a snippet of our conversation. For the entire interview, follow along at our Innovations Projects blog.

To start, let’s talk about the first few months of Solo since its Beta launch in December. How big is the team, and how many websites have been published?

Solo started in May of last year with just myself and a part-time designer. We built a lightweight, clickable prototype and technical implementation to test whether generative AI could really work for website authoring. We also spent significant time surveying the broader landscape to make sure we had something that could differentiate and compete in an entrenched market.

After initial validation, we added one engineer and started the development process in June. By September, we had our first iteration that could create a website for a solopreneur with just a few simple inputs, and from there, we continued to refine the user experience. In October, we released an internal beta, and then in December, we launched our beta publicly.

Since then, our team within the Mozilla Innovation Projects group grew during this journey from just an engineer, a part-time designer and me, to three engineers, a part-time designer, myself and other part-time resources to support us.

We launched Solo 1.0 this past month, and in that period since beta, we’ve seen over 7,000 published websites across industries, from pool cleaners, to coaches, to immigration consultants.

When we started, our goal was to make it simple for non-technical solopreneurs to build their web presence and grow their business, and we believe we have accomplished the first step.

How do you compete with Solo in such a crowded market? How are you making Solo free?

When we initially conceptualized Solo, marrying GenAI with the service provider segment was an insertion point. Since then, and as expected, the incumbents have also built GenAI capabilities and improved their user experiences for the service provider audience.

In this situation, where we are the underdog, my approach is to look for maximum disruption, and we landed at the business model. Every competitor — that we are aware of, at least — charges for connecting and hosting your custom domain. This makes sense – 20 years ago, bandwidth wasn’t cheap and SSL (Secure Sockets Layer) certificates that enable an encrypted web connection cost money. Today, the former is near zero and the latter is zero.

We asked, “What if we just made this free?” This would be very disruptive, so this is exactly what we are doing. Not only is it disruptive, but democratizing the category is also in line with our mission to increase access to the web. We are making web hosting and connecting your custom domain free, similar to how Robinhood disrupted brokerages by eliminating trading fees. Many do not launch their website because they can’t afford or don’t yet have enough business to justify the cost. It also doesn’t help that many of these incumbents rely on hidden upsells and next thing you know, you’re spending $100 a month for your dog walking service. In this way, Mozilla continues to be a global public resource looking out for the interests of people.

How is it building Solo, a new product at Mozilla?

Building a new product at Mozilla, also known as zero to one, and probably any large organization, has challenges. I come from a startup background, so this is my comfort zone and I have some principles I generally abide by.

First, it’s important to be the top advocate for the product. This can be hard because things will pull you in different directions, whether other initiatives, shiny objects or your own self-doubt. Second, adopt the tools you need and optimize on speed — it’s easy to get stuck in administrative stuff. Third, resourcing can be slow, so optimize on generalists and make sure everyone is comfortable with grunt work. Fourth, make product decisions — many of them one-way — quickly. There’s just not enough time to get consensus or have everything be data-driven at the onset. The last thing is to take agency when you can. The five minutes here, the half day there, the follow-up meeting tomorrow cause delays and they compound.

Can you share details about how Solo fits into Mozilla’s overall mission?

In many ways, Mozilla has been at the intersection of the internet and the interests of the people, as opposed to big tech. With Solo, we are squarely within this vision. We are democratizing access to the web for solopreneurs, and we’re increasing equity by helping those that can’t afford to host their websites in emerging markets, or where English isn’t their first language, with writing, designing and curating their content.

For the entire interview, follow along at our Innovations Projects blog.

Ready to start creating?

Launch your website

The post How Mozilla’s AI website creator, Solo, is shaking up a $2.1B industry appeared first on The Mozilla Blog.

Spotting unreliable reviews on TikTok’s trending products with Fakespot

Fakespot is a browser extension that is powered by AI to help millions of shoppers make better purchases. It analyzes product reviews in real-time across supported retailers like Amazon, Best Buy, and Walmart, giving you the lowdown on which product reviews seem credible and when you should proceed with caution. Whether you’re seeking product pros and cons or highlights, Fakespot has you covered. It even provides seller ratings for eBay and Shopify stores so you can shop with confidence.

How to read Fakespot’s Review Grades

Here’s how Fakespot’s grading system works:

• A and B: Reviews you can trust.
• C: Mixed bag of reliable and unreliable reviews — approach with caution.
• D and F: Probably unreliable.

Top TikTok back-to-school product categories on Amazon and their reliability

We took a deep dive into top back-to-school categories and analyzed their Fakespot product review grades. Here’s what we found:

Laptops

Impressively, nearly 83% of laptop reviews on Amazon appear to be reliable. A safe bet, especially if you’re shopping with the Fakespot extension and sticking to verified retailers. The Lenovo Yoga 7i, a popular choice on TikTok, gets a Fakespot review grade of A. 

• Trending Product: Lenovo Yoga 7i
• Fakespot Review Grade: A 
• Review highlight: “For the price, you can’t beat it.” 

Water bottles

About 30% of water bottle reviews on Amazon appear unreliable. Staying hydrated? Just make sure you’re buying from verified retailers. The Thermos Hydration Bottle has been trending on TikTok and gets a B grade from Fakespot, meaning the product listing has pretty reliable reviews. 

• Trending Product: Thermos Hydration Bottle
• Fakespot Review Grade: B
• Review highlight: “Easy to grip, fits in car cupholders.”

Keyboards and mice

While 34% of Amazon reviews in the keyboards and mice category appear to be unreliable, one popular gaming mouse still earns a solid B from Fakespot, so you can be more confident in the reviews. 

• Trending Product: Razer DeathAdder V3 Pro Wireless Gaming Mouse
• Fakespot Review Grade: B
• Review highlight: “The razer deathadder v3 pro wireless gaming mouse ‘faker edition’ is a tribute to all gamers who seek excellence. It eliminates the shackles of wired gaming, allowing you the freedom to move and game as you wish without compromising on responsiveness or speed.”

Pillows

With 37% of Amazon reviews on pillows appearing unreliable, those trendy pillows still have potential. If you’re looking for a decorative dorm pillow, we found an option with reliable reviews for you to consider. 

• Trending Product: Wedge Body Pillow 
• Fakespot Review Grade: A
• Review highlight: “The item is well-made, very soft and comfortable. Even has a place to hold his cell phone.”

Backpacks

Fakespot detected concerns with nearly half – 47% – of the reviews on backpacks on Amazon. Despite this, we found a durable option with an A review rating from Fakespot. Nonetheless, before you snag any bag, let Fakespot give you confidence that those raves are reliable.

• Trending Product: Laptop Backpack 
• Fakespot Review Grade: A
• Review highlight: “It was light but sturdy and has many pockets for storage.”

Chargers

Those fast chargers blowing up on TikTok? They might not be so fast after all. About 53% of Amazon reviews for chargers seem unreliable. However, this listing for a popular wireless charging dock gets an A from Fakespot, so you can be more confident in the reviews.

• Trending Product: Belkin 2-in-1 MagSafe Wireless Charging Dock
• Fakespot Review Grade: A
• Review highlight: “The base is heavy enough that you don’t pull it around while taking your phone off the charger — it feels like quality material.”

Earbud headphones and computer accessories

In the earbud, headphones and computer accessories category on Amazon, more than half appear to be unreliable reviews (58%)  – so take a beat before you buy. But these noise-canceling headphones, a TikTok favorite, still get a B grade from Fakespot, meaning the reviews are more reliable. 

• Trending Product: Sony WH-1000XM4
• Fakespot Review Grade: B
• Review highlight: “I love blasting my music. I enjoy barbershop music, and you can hear each individual voice so very well in great quality. I’m stunned.”

If you’re one of the millions of students gearing up for the new school year, don’t waste time scrolling through endless product reviews. Download Fakespot today, and spend your time soaking up the last bits of summer instead.

Shop confidently with Fakespot

Download the latest version today

The post Fakespot’s guide to trending back-to-school products  appeared first on The Mozilla Blog.

This past year has been marked by more and more people recognizing the societal benefits of open source AI. In October, a large coalition of people signed onto our statement emphasizing how openness and transparency are critical ingredients to safety and security in AI. In February, Mozilla and the Columbia Institute of Global Politics convened AI experts, who emphasized how openness in AI could help advance key societal goals. Policymakers have also been embracing open source AI. The U.S. National Telecommunications and Information Administration (NTIA) recently issued a seminal report embracing openness in AI. Even companies like Google, Microsoft, Apple, and Meta are beginning to open certain aspects of their AI systems.

The growing focus on open source AI makes it all the more important that we establish a shared understanding of what open source AI is. A definition should outline what must be shared and under what terms or conditions. Without this clarity, we risk a fragmented approach, where companies label their products as “open source” even when they aren’t, where civil society doesn’t have access to the AI components they need for testing and accountability, and policymakers create regulations that fail to address the complexities of the issue.

The Open Source Initiative (OSI) has recently released a new draft definition of open source AI, marking a critical juncture in the evolution of the internet. This moment comes after two years of conversations, debates, engagements, and late-night conversations across the technical and open source communities. It is critical not just for redefining what “open source” means in the context of AI; it’s about shaping the future of the technology and its impact on society.

The original Open Source Definition, introduced by the OSI in 1998, was more than just a set of guidelines; it was a manifesto for a new way of building software. This definition laid the foundation for open systems that have since become the backbone of the modern internet. From Linux to Apache, open source projects have driven innovation, collaboration and competition, enabling the internet to grow into a diverse and dynamic ecosystem. By ensuring that software could be freely used, modified, and shared, the original open source movement helped to expand access to technology, breaking down barriers to entry and fostering a culture of innovation and transparency, while making software safer and less vulnerable to cyberattacks.

This is a significant step toward bringing clarity and rigor to the open source AI discussion. It introduces a binary definition of “open source,” akin to the existing definition. While this is just one of several approaches to defining open source AI, it offers precision for developers, advocates and regulators who benefit from clear definitions in various working contexts. Specifically, it outlines that open source AI revolves around the ability to freely use, study, modify and share an AI system. And it also promotes the importance of access to key components needed to recreate substantially equivalent AI systems, like information on data used for training, the source code for AI development and the AI model itself.

And, this definition also offers an initial attempt to wrestle with the complex issue of whether and how training data for AI models should be shared as part of open source AI. The definition acknowledges that sharing full training datasets can be challenging in practice, and, therefore, avoids disqualifying a significant amount of otherwise open source AI development from being considered “open source.” We are working to change this state of play by making open datasets a more commonplace part of the AI ecosystem. Mozilla and Eleuther AI recently brought together experts to outline best practices for open datasets to support AI training, and we intend to publish a paper soon that promotes norms that support AI training data being more widely available.

We acknowledge that some may disagree with aspects of OSI’s definition, such as its treatment of training data, and that the definition will need refinement over time. However, we believe that the OSI’s community-driven process — which involved over a year of stakeholder engagement — has established a crucial reference point for discussions on open source AI. For instance, this definition will become a valuable resource to combat the widespread practice of “openwashing” that is becoming quite rampant, where non-open models (or even open-ish models like Meta’s Llama 3) are promoted as leading “open source” options without contributing to the commons. Researchers have shown that “the consequences of open-washing are considerable” and affect innovation, research and the public understanding of AI.

At its core, this effort embodies the open source community at its best — engaging in open discussions, addressing differences, acknowledging shortcomings and refining this definition together, to build something better. It effectively incorporates many key aspects of openness that the open source community has been grappling with, such as going beyond just considering openness in model weights and including broader model components, documentation, and licensing approaches as outlined in the Columbia Convening. In contrast, the closed source ecosystem operates in secrecy, with limited access and behind-the-scenes deals where large tech companies exchange compute power and talent. We prefer our sometimes imperfect but consistently transparent approach any day.

We, and many others, are eager to continue collaborating with OSI and the broader open source community to bring greater clarity to the open source AI discussion and continue unlocking the potential of open source AI for the benefit of society.

Get Firefox

Get the browser that protects what’s important

The post Celebrating an important step forward for open source AI appeared first on The Mozilla Blog.

The second annual Rise25 Awards in Dublin wasn’t just about celebrating 25 AI leaders. It was about mapping out the future.

The Gardiner Brothers, known for bringing Irish dance into the social media spotlight, kicked off the night with a performance that mixed tradition and innovation. Siobhán McSweeney of “Derry Girls” fame hosted the ceremony, and she kept the crowd engaged with humor, quipping, “AI touches everything like a child with sticky fingers that comes around to the house, just after you cleaned it.”

Then, the honorees took the stage to lay out the principles guiding their work. Here are the highlights:

The stories we tell about AI shape its future

Sinéad Bovell, one of the 2024 Rise25 honorees in the artist category, works on preparing young people for a future driven by advanced tech. She emphasized that the narratives we craft around AI are crucial; they frame public understanding and ultimately influence the direction of AI development. 

“It’s such an honor to be recognized in the artist category, because the stories that we tell about artificial intelligence matter deeply,” Sinéad said. She pointed out that it’s easy to feel trapped in a binary narrative about AI, with dangers and risks on one side and benefits and possibilities on the other. “But the truth is, these stories aren’t separate. They’re intertwined,” she said. 

Citing technologist Jaron Lanier, Sinéad argued that to be a true optimist about AI, you also need to be a fierce critic. “We have to continue to tell the stories of a future where we get AI right and where it transforms humanity for the better. But we also have to tell the stories of how we got there, the challenging decisions we made in the present, and where we chose to keep humanity at the center of technological advancements.” For Sinéad, understanding and telling these nuanced stories is essential for guiding AI toward an ethical and inclusive future.

AI’s effects on individuals can be profound

Gemma Galdon-Clavell, an honoree in the entrepreneur category, is focused on finding and fixing bias and flaws in predictive and large language model (LLM) tools. She shared a deeply personal story that underscored the far-reaching impact AI can have on individuals:

“If my school had used an AI system to assess my chances, I wouldn’t be here today. My mom was 14 when she had me. I had huge behavioral problems growing up. If you had inserted all that data into an AI system and asked, ‘Should this girl go to school? Should we invest in her?’ The answer would have been no.” 

Gemma highlighted the dangers of relying solely on algorithms to determine someone’s potential, as these systems often reduce complex lives to mere data points. “I am here because I managed to beat the odds — because no one set my odds in an algorithm.” 

Her story serves as a powerful reminder of the need for rigorous oversight and auditing of AI systems to ensure they don’t limit the futures of those who, like her, might defy expectations. “People, like everyone else, deserve a chance,” she concluded, advocating for a future where AI supports rather than stifles human potential.

Design choices in AI have far-reaching consequences

Philosopher and AI Ethics Lab founder Cansu Canca, a change agent honoree, spoke passionately about the critical importance of ethical design in AI, highlighting how every decision made in the design process has the potential to shape society. 

“When we design AI systems, we’re not just making technical choices. We’re making moral and philosophical decisions,” she said. Cansu challenged developers to consider questions that go beyond code: “What is a good life? What is a better society?” These questions, she argued, should guide every step of AI development. 

“The design choices we make today will determine whether AI becomes a tool for justice or a mechanism that perpetuates inequality,” Cansu warned. She called for an approach to AI that integrates ethical considerations from the outset, ensuring that systems are designed to promote fairness, transparency and respect for human dignity. “Ethical design isn’t an afterthought — it’s the foundation on which AI should be built,” she said, stressing the far-reaching impact of these decisions on our collective future.

AI’s untapped potential lies in open collaboration

Researcher Aaron Gokaslan, an honoree in the builder category, aims to keep generative model development open. He highlighted the immense, largely untapped potential of AI, particularly within the realm of open-source development. 

“We’re in the very early innings of AI today,” he remarked, pointing out that while AI has already made significant strides, its full potential is still on the horizon. Aaron emphasized that the true power of AI will be unlocked through collaboration and accessibility, which would enable a diverse range of innovators to contribute to its development. 

“By sharing knowledge and resources, we can drive AI forward in ways that benefit society as a whole,” Aaron said.

Thoughtful AI policy is essential for a fair future

Philip Thigo, an advocate honoree and the special envoy on technology for the Republic of Kenya, underscored the critical need for thoughtful and proactive AI policy, warning that without it, AI could deepen existing inequalities and erode public trust. 

“AI has the potential to revolutionize society, but without robust and thoughtful regulation, it could also exacerbate inequalities and undermine public trust,” he cautioned. Philip argued that AI policy must prioritize fairness, transparency and accountability to ensure that AI development benefits everyone, not just a privileged few. 

“We need policies that don’t just react to AI’s challenges, but anticipate them — setting clear guidelines for ethical development and use,” he said. Philip called for a collaborative approach to AI governance, involving not only policymakers and technologists, but also the broader public: “By engaging a diverse range of stakeholders, we can create a framework that guides AI toward serving the common good.”

As the honorees made clear, AI will leave its mark wherever it touches, much like those “sticky fingers.” The challenge ahead is making sure that mark is a positive one. The evening was a powerful reminder that the future of AI is not just about innovation — but about inclusivity, ethics and accountability. 

Get Firefox

Get the browser that protects what’s important

The post At the Rise25 Awards, the future of AI is ethical, inclusive and accountable appeared first on The Mozilla Blog.

“I’m so excited to host these awards. AI is one of the biggest issues facing us, not only in my industry but across the board. To recognise and award individuals who are working for the benefit of society and not corporations is a great honor,” said McSweeney. She continued: “I’m so looking forward to meeting them. And having them explain what AI is.”

Rise25 is more than an award ceremony—it’s a platform to spark discussion, forge connections and inspire a wave of new ideas that will shape the future of AI. Siobhán’s contributions to television and theater, characterized by depth and charisma, make her an ideal voice to help us highlight these themes.

“The Rise25 awards are committed to bridging the gap between complex technological innovations and the very human stories at their core,” said Mark Surman, President of Mozilla. “Ms. McSweeney’s portrayal of Sister Michael in ’Derry Girls’ has left a lasting impression on audiences around the world (we’re big fans!). Her ability to deliver lines with sharp wit, while maintaining a warm presence, perfectly encapsulates the blend of insight and approachability we covet at our event.”

The awards show will also feature a special performance by Galway, Ireland-based Irish dancers The Gardiner Brothers. The American-born Irish dancers have won over 40 major Irish dancing titles between them and have performed to audiences all over the globe, including with the world famous Riverdance cast. They are known for their fast paced and rhythmic style of dance that they developed after training at the Hession School of Irish Dance in Galway, Ireland.

Bridget Todd, host of Mozilla’s Webby Award-winning “IRL” podcast, will be on hand to present the award categories at this year’s ceremony. Bridget is also the host of the iHeart Radio Podcast Award-winning podcast “There Are No Girls On The Internet,” and is a Shorty Award winner for “Best Podcast Miniseries” for DISINFORMED, a miniseries exploring how misinformation, and conspiracy theories around COVID, gender, and race hurt marginalized communities. Bridget’s writing and work on technology, race, gender and culture have been featured at The Atlantic, Newsweek, “The Daily Show” and more.

Mozilla’s 2nd Annual Rise25 Awards build upon the success of last year’s Rise25 Awards which were held in Berlin, Germany, bringing to life what a future trustworthy Internet could look like.

This year’s awards ceremony will be available on demand on Mozilla’s YouTube channel the morning of Friday, August 16, 2024. For more information, please visit https://rise25.mozilla.org/ 

The post BAFTA Award-Winner Siobhán McSweeney to host Mozilla’s 2nd Annual Rise25 Awards in Dublin, Ireland on Aug. 13 appeared first on The Mozilla Blog.

At Mozilla, we work hard to make Firefox the best browser for you. That’s why we’re always focused on building a browser that empowers you to choose your own path, that gives you the freedom to explore without worry or compromises. We’re excited to share more about the updates and improvements we have in store for you over the next year.

Bringing you the features you’ve been asking for

We’ve been listening to your feedback, and we’re prioritizing the features you want most.

• Productivity boosters like
  • Tab Grouping, Vertical Tabs, and our handy Sidebar will help you stay organized no matter how many tabs you have open — whether it’s 7 or 7,500. 
  • Plus, our new Profile Management system will help keep your school, work, and personal browsing separate but easily accessible. 
• Customizable new tab wallpapers that will let you choose from a diverse range of photography, colors, and abstract images that suits you most. 
• Intuitive privacy settings that deliver all the power of our world-class anti-tracking technologies in a simplified, easy-to-understand way.
• More streamlined menus that reduce visual clutter and prioritize top user actions so you can get to the important things quicker.

Continuous work on speed, performance and compatibility

Speed is everything when you’re online, so we’re continuing to work hard to make Firefox as fast and efficient as possible. You can expect even faster, smoother browsing on Firefox, thanks to quicker page loads and startup times – all while saving more of your phone’s battery life. We’ve already improved responsiveness by 20 percent as measured by Speedometer 3, a collaboration we’ve spearheaded with other leading tech companies. And in that collaborative spirit, we’re also working with the Interop project to make it easy for people to build sites that work great across all browsers. We value your support in our efforts to improve cross-browser compatibility which is why we’ve added new features to easily report when websites aren’t working quite right; this feedback is critical as we look to address even small functionality issues that affect your day-to-day online experience.

Making the most of your time online — without sacrifice

Ensuring your privacy is core to everything we do at Firefox. Unlike other companies, who ask you to exchange your data in order to do even basic, everyday things online — you don’t have to give up your personal information to get a faster, more efficient browser experience with Firefox. Reading a news story in a different language or signing a form for school or work shouldn’t require you to give up your privacy. So, we’ve worked hard to make things like translation and PDF editing in Firefox happen locally on your device, so you don’t have to ship off your personal data to a server farm for a company to use it how they see fit — to keep tabs on you, sell your information to the highest bidder, or train their AI. With Firefox, you have a lot of choice — but you don’t have to choose between utility and privacy. Your data is secure, and most importantly, just yours.

We are approaching the use of AI in Firefox — which many, many of you have been asking about — in the same way. We’re focused on giving you AI features that solve tangible problems, respect your privacy, and give you real choice.

We’re looking at how we can use local, on-device AI models — i.e., more private — to enhance your browsing experience further. One feature we’re starting with next quarter is AI-generated alt-text for images inserted into PDFs, which makes it more accessible to visually impaired users and people with learning disabilities.

Join us on this journey

Our progress is driven by a vibrant community of users and developers like you. We encourage you to contribute to our open-source projects and to engage with us on Mozilla Connect or Discourse, and check out our recent AMA on Reddit. Your participation is crucial in shaping what Firefox becomes next.

Get Firefox

Get the browser that protects what’s important

The post Here’s what we’re working on in Firefox appeared first on The Mozilla Blog.

“We’re all about helping you shop smarter, especially during this month’s summer sales and the upcoming back-to-school season,” said Saoud Khalifah, co-founder and director of Fakespot. “Our latest report shows just how crucial it is to check those reviews, especially in categories flooded with unreliable reviews. By spotlighting both the best and worst categories, we give you the tools to shop with confidence.”

Since 2016, Fakespot has empowered millions of shoppers to make well-informed purchases using advanced AI technology. Its AI engine analyzes reviews, filters out unreliable ones and gives shoppers a true understanding of the quality of a product and the seller, so they can feel confident about their decisions. As a free browser extension available on most web browsers, Fakespot analyzes reviews from top e-commerce sites like Amazon, Best Buy, Sephora and Walmart, providing the most reliable product information before you buy. Bonus: The extension also provides seller ratings on Shopify-powered web stores.

When it comes to online shopping, knowing which products are trustworthy can save time and money. We are sharing our latest findings just in time for the shopping season. Our latest analysis has revealed some surprising and noteworthy stats (June 1, 2023 through May 31, 2024). Here are the most reliable and least reliable popular product categories:

Shop with confidence: Top 5 categories you can trust 

Shop confidently with these top-rated products. These categories earned Fakespot Grades of B or better for reliability.

1. Apple products: With an impressive 84% of reviews being genuine, Apple products stand out as a top choice for reliability. Only a tiny 5% of reviews are marked as unreliable.
2. Video game chairs: Gamers, rejoice! 84% of reviews for video game chairs are trustworthy, making this a solid category for your next purchase.
3. Books: Book lovers can breathe easy, as 81% of book reviews are authentic. Even with a whopping 2,907 products reviewed, books maintain a high standard of reliability.
4. Computers: Tech enthusiasts, take note. Computers come in with 79% genuine reviews, ensuring you get the real scoop before buying.
5. Home Office Desks: Perfect for remote work and homework, with 68% of reviews being legit.

Shop carefully: Top 5 categories to watch out for 

Shop carefully and think twice before purchasing from these categories. These product categories have a Fakespot Review Grade of D or lower.

1. Slides: A staggering 75% of reviews for slides are unreliable, making it the least reliable category in our study.
2. Pajamas: Cozy up with caution, as 62% of pajama reviews aren’t genuine.
3. Basketball: Sports gear shoppers should be wary, with 61% of basketball-related product reviews being unreliable.
4. Stick vacuums and electric brooms: Housekeeping might need a bit more homework, with 57% unreliable reviews in this category.
5. Fashion hoodies and sweatshirts: Fashion fans, beware. Over half (57%) of reviews in this category are not reliable, despite the large number of products reviewed (6,078).

Millions of Fakespot users depend on Fakespot’s Review Grade to help determine the reliability of the product reviews and seller. It follows the standard grading system of “A”, “B”, “C”, “D”, or “F” and represents the following:

• Fakespot Review Grade A and B: These grades represent reliable reviews.
• Fakespot Review Grade C: This grade should be approached with caution, as it includes a mix of reliable and unreliable reviews.
• Fakespot Review Grade D and F: These grades are considered unreliable.

We know just how crucial reliable reviews are in making informed purchasing decisions. Fakespot’s study sheds light on which categories are more prone to review manipulation, helping consumers make smarter, more informed choices.

So, whether it’s deal days or early back-to-school shopping, be sure to download Fakespot whenever you shop online. 

Shop confidently with Fakespot. Download the latest version today.

The post Fakespot reveals the product categories with the most and least reliable product reviews appeared first on The Mozilla Blog.

Today, U.S. Senator Maria Cantwell (D-Wash.), Chair of the Senate Committee on Commerce, Science and Transportation, convened a full committee hearing titled “The Need to Protect Americans’ Privacy and the AI Accelerant.” The hearing explored how AI has intensified the need for a federal comprehensive privacy law that protects individual privacy and sets clear guidelines for businesses as they develop and deploy AI systems. 

Mozilla’s Director of Global Product Policy, Udbhav Tiwari, served as a key witness at the public hearing, highlighting privacy’s role as a critical component of AI policy. 

“At Mozilla, we believe that comprehensive privacy legislation is foundational to any sound AI framework,” Tiwari said. “Without such legislation, we risk a ‘race to the bottom’ where companies compete by exploiting personal data rather than safeguarding it. Maintaining U.S. leadership in AI requires America to lead on privacy and user rights.” Tiwari added that data minimization should be at the core of these policies.

As a champion of the open internet, Mozilla has been committed to advancing trustworthy AI for half a decade. “We are dedicated to advancing privacy-preserving AI and advocating for policies that promote innovation while safeguarding individual rights,” Tiwari said. 

Read the written testimony. 

The post Mozilla heads to Capitol Hill, calls for a federal privacy law to ensure the responsible development of AI appeared first on The Mozilla Blog.

Recently, we shared our 2024 priorities for Firefox on Mozilla Connect, our dedicated space for engaging our community. It’s where we exchange ideas, gather feedback and spark discussions to shape future product releases. Eager to reach more of the community, the Firefox team hosted a two hour Ask Me Anything on Reddit.

The AMA had a great turnout of Firefox supporters keen to discuss topics like videos for new releases, performance enhancements and feature suggestions like automatic tab discarding. Here are some highlights: 

User interface and productivity enhancements 

Our community is enthusiastic about several upcoming features designed to improve productivity and customization. Our Tab Groups team is hard at work creating a better experience for people who need to manage many tabs. Additionally, we are also exploring modernizing a place where you’ll have access to better tools for viewing your browser history and bookmarks and providing a better interface for organizing tabs and windows across devices.

Performance and compatibility improvements 

Speed and efficiency remain top priorities. For Android users, enhancements focus on improving performance, battery life and adding new features. Site isolation (Fission) is also in progress for Android to boost security.

Media and accessibility features 

To enhance media playback, we’re exploring a contributor patch — a piece of code submitted by an external developer — for Picture-in-Picture (PiP) autoplay, a feature that allows video content to play in a small, resizable window that stays on top of other windows, which will automatically open a Picture-in-Picture window when navigating away from a media tab. On the accessibility front, we’re expanding our translation feature to include more languages, with new packages and support for Chinese, Japanese, and Korean in our pipeline.

We want to hear from you 

This was our first AMA in a while, and we’re interested in what you think. Did you find the AMA helpful? Was there anything we missed? How often should we host these sessions? Let us know by filling out this survey. If you missed the AMA, you can still check it out here.

These insights and feedback from our community are instrumental in guiding our efforts to make Firefox the best browser experience possible. Stay tuned for more on our next AMA as well as updated. Join us on Mozilla Connect to keep the conversation going!

Get Firefox

Get the browser that protects what’s important

The post Engaging with our Firefox community: Highlights from our recent AMA  appeared first on The Mozilla Blog.

Mozilla and EleutherAI brought together experts to discuss a critical question: How do we create openly licensed and open-access LLM training datasets and how do we tackle the challenges faced by their builders?

On June 11, on the eve of MozFest House in Amsterdam, Mozilla and EleutherAI convened an exclusive group of 30 leading scholars and practitioners from prominent open-source AI startups, nonprofit AI labs and civil society organizations to discuss emerging practices for a new focus within the open LLM community: creating open-access and openly licensed LLM training datasets. 

This work is timely. Although sharing training datasets was once common practice among many AI actors, increased competitive pressures and legal risks have made it almost unheard of nowadays for pre-training datasets to be shared or even described by their developers. However, just as open-source software has made the internet safer and more robust, we at Mozilla and EleutherAI believe open-access data is a public good that can empower developers worldwide to build upon each other’s work. It fosters competition, innovation and transparency, providing clarity around legal standing and an ability to stand up to scrutiny.

Leading AI companies want us to believe that training performant LLMs without copyrighted material is impossible. We refuse to believe this. An emerging ecosystem of open LLM developers have created LLM training datasets —such as Common Corpus, YouTube-Commons, Fine Web, Dolma, Aya, Red Pajama and many more—that could provide blueprints for more transparent and responsible AI progress. We were excited to invite many of them to join us in Amsterdam for a series of discussions about the challenges and opportunities of building an alternative to the current status quo that is open, legally compliant and just. 
During the event, we drew on the learnings from assembling “Common Pile” (the soon-to-be-released dataset by EleutherAI composed only of openly licensed and public domain data) which incorporates many learnings from its hugely successful predecessor, “The Pile.” At the event, EleutherAI released a technical briefing and an invitation to public consultation on Common Pile.

Our goal with the convening was to bring in the experiences of open dataset builders to develop normative and technical recommendations and best practices around openly licensed and open-access datasets. Below are some highlights of our discussion:

• Openness alone does not guarantee legal compliance or ethical outcomes, we asked which decision points can contribute to datasets being more just and sustainable in terms of public good and data rights. 
• We discussed what “good” looks like, what we want to avoid, what is realistic and what is already being implemented in the realm of sourcing, curating, governing and releasing open training datasets. 
• Issues such as the cumbersome nature of sourcing public domain and openly licensed data (e.g. extracting text from PDFs), manual verification of metadata, legal status of data across jurisdictions, retractability of consent, preference signaling, reproducibility and data curation and filtering were recurring themes in almost every discussion.
• To enable more builders to develop open datasets and unblock the ecosystem, we need financial sustainability and smart infrastructural investments that can unblock the ecosystem.
• The challenges faced by open datasets today bear a resemblance to those encountered in the early days of open source software (data quality, standardization and sustainability). Back then, it was the common artifacts that united the community and provided some shared understanding and language. We saw the Dataset Convening as an opportunity to start exactly there and create shared reference points that, even if not perfect, will guide us in a common direction.
• The final insight round underscored that we have much to learn from each other: we are still in the early days of solving this immense challenge, and this nascent community needs to collaborate and think in radical and bold ways.

We are immensely grateful to the participants in the Dataset Convening (including some remote contributors):

• Stefan Baack — Researcher and Data Analyst, Insights, Mozilla
• Mitchell Baker — Chairwoman, Mozilla Foundation
• Ayah Bdeir — Senior Advisor, Mozilla
• Julie Belião — Senior Director of Product Innovation, Mozilla.ai
• Jillian Bommarito — Chief Risk Officer, 273 Ventures
• Kasia Chmielinski — Project Lead, Data Nutrition Project
• Jennifer Ding — Senior Researcher, Alan Turing Institute
• Alix Dunn — CEO, Computer Says Maybe
• Marzieh Fadaee — Senior Research Scientist, Cohere For AI
• Maximilian Gahntz — AI Policy Lead, Mozilla
• Paul Keller — Director of Policy and Co-Founder, Open Future
• Hynek Kydlíček — Machine Learning Engineer, HuggingFace
• Pierre-Carl Langlais — Co-Founder, Pleias
• Greg Leppert — Director of Product and Research, the Library Innovation Lab, Harvard
• EM Lewis-Jong — Director, Common Voice, Mozilla
• Shayne Longpre — Project Lead, Data Provenance Initiative
• Angela Lungati — Executive Director, Ushahidi
• Sebastian Majstorovic — Open Data Specialist, EleutherAI
• Cullen Miller — Vice President of Policy, Spawning
• Victor Miller — Senior Product Manager, LLM360
• Kasia Odrozek — Director, Insights, Mozilla
• Guilherme Penedo — Machine Learning Research Engineer, HuggingFace
• Neha Ravella — Research Project Manager, Insights Mozilla
• Michael Running Wolf — Co-Founder and Lead Architect, First Languages AI Reality, Mila
• Max Ryabinin — Distinguished Research Scientist, Together AI 
• Kat Siminyu — Researcher, The Distributed AI Research Institute
• Aviya Skowron — Head of Policy and Ethics, EleutherAI
• Andrew Strait — Associate Director, Ada Lovelace Institute
• Mark Surman — President, Mozilla Foundation
• Anna Tumadóttir — CEO, Creative Commons
• Marteen Van Segbroeck — Head of Applied Science, Gretel
• Leandro von Werra — Chief Loss Officer, HuggingFace
• Maurice Weber — AI Researcher, Together AI
• Lee White — Senior Full Stack Developer, Ushahidi
• Thomas Wolf — Chief Science Officer and Co-Founder, HuggingFace

In the coming weeks, we will be working with the participants to develop common artifacts that will be released to the community, along with an accompanying paper. These resources will help researchers and practitioners navigate the definitional and executional complexities of advancing open-access and openly licensed datasets and strengthen the sense of community. 

The event was part of the Mozilla Convening Series, where we bring together leading innovators in open source AI to tackle thorny issues and help move the community and movement forward. Our first convening was the Columbia Convening where we invited 40 leading scholars and practitioners to develop a framework for defining what openness means in AI. We are committed to continuing the efforts to support communities invested in openness around AI and look forward to helping grow and strengthen this movement. 

The post The Dataset Convening: A community workshop on open AI datasets appeared first on The Mozilla Blog.