Kalakshar US

One question is constant in the race to create AI: Should code be held privately or made available to everyone as “open-source”? OpenAI has kept data for projects like GPT-4 close to its chest, while Meta has allowed researchers and academics to access its language model, LLaMA. Now, Meta is allegedly set to “imminently” release a new commercial version that companies can customize, the Financial Times reports.

The move could help Meta catch up to fellow AI creators OpenAI and Google, with businesses able to build tailored software with the new model. “The competitive landscape of AI is going to completely change in the coming months, in the coming weeks maybe, when there will be open source platforms that are actually as good as the ones that are not,” Yann LeCun, Meta’s vice-president and Chief AI scientist, said at a July conference. Meta might also create AI chatbots specific to each type of user, such as an individual or business.

Open-source models certainly have a mix of benefits and negatives. Providing a wide group of people with an AI’s differing capabilities can help the model learn faster, given the increased data it’s receiving. Additional eyes can also spot bugs and security issues, giving developers a chance to fix these problems sooner. On the flip side, not everyone who uses the code will have good intentions, which can have dangerous consequences when using a technology that concerns many people in and outside the tech industry.

Access to Meta’s commercial AI model should be free at launch, but that might change sometime in the future. According to sources familiar with Meta, the company might eventually charge enterprise customers if they want to alter or tailor the model.

This article originally appeared on Engadget at https://www.engadget.com/meta-could-soon-make-its-ai-model-available-for-commercial-projects-114021749.html?src=rss 

Apple has expanded Tap to Pay’s availability and is giving businesses and merchants across the United Kingdom the option to accept contactless payments using just their iPhones. The tech giant initially released the feature in the US in mid-2022 before making it available in Taiwan and Australia. Like in those regions, merchants in the UK will be able to receive payments without having to purchase extra hardware, since the capability uses NFC. As long as they have an iPhone XS or later, they can offer contactless payment as an option for customers who use Apple Pay, contactless debit or credit cards and other digital wallets. 

The company unveiled the feature in early 2022 before testing the payment system at its own visitor center in Cupertino. Following Tap to Pay’s rollout, big-name fintech services launched support for the feature, including PayPal, Venmo and Square. 

According to Apple’s announcement, Revolut and Tyl by NatWest are the first payment platforms allowing their business customers in the UK to use the company’s Tap to Pay technology. Adyen, Dojo, myPOS, Stripe, SumUp, Viva Wallet, Worldline and Zettle by PayPal will be the next financial services to support the feature in the region. Apple didn’t mention a specific rollout date for them, only that support for Tap to Pay will be “coming soon” for their customers. In addition announcing the feature’s availability, the company has revealed that Tap to Pay will be available in its retail stores across the UK in the coming weeks. That means UK customers will be easily be able to pay for purchases by bumping a card or another phone to the store’s iPhone.

This article originally appeared on Engadget at https://www.engadget.com/apple-tap-to-pay-expands-to-the-uk-letting-iphone-users-accept-contactless-payments-101724510.html?src=rss 

Google’s Bard gained a handful of new features and functions Thursday in the chatbot AI’s latest round of updates, including expanded linguistic knowledge, more nuanced response controls and the ability to respond with spoken word in addition to text. In all the AI can now converse in nearly four dozen languages. 

Users can now converse with the AI in Arabic, Chinese, German, Hindi and Spanish, among others as well as access the platform from more places on the planet, such as Brazil and “across Europe,” Jack Krawczyk, Bard Product Lead, and Amarnag Subramanya, Bard’s VP of Engineering, wrote in a blog post Thursday. “As we bring Bard to more regions and languages over time, we’ll continue to use our AI Principles as a guide, incorporate user feedback, and take steps to protect people’s privacy and data.”

Bard now literally speaks. Users will have the option to either read or listen to the AI’s generated responses, which Krawczyk and Subramanya believe will help immensely when users want to hear the correct pronunciation of words in those 40 newly-added languages. Users have also been afforded more robust controls over how friendly Bard will be with five distinct options for the AI’s tone: simple, long, short, professional or casual. Those are only available on English-language requests for the moment but the company is already working to expand it out to more of the 40, “soon.”

The chatbot also has some fancy new multimodal eyes, gaining the capacity to interpret images dropped into the chat through the prompt field. Faster and easier than uploading it as a document, users can request more information about the contents of the image or generate content like captions based on it. This also is currently English-only. 

Getting the information and code that Bard generates out of the chat window and into the hands of collaborators is no longer quite such a slog. Starting Thursday, users will be able to export Bard-generated Python code to Replit, in addition to Colab. They’ll also be able to copy and share portions of individual chats with other users. The process of organizing and revisiting old conversations being streamlined as well with the addition of pinned conversations, which are what they sound like, and the ability to rename them.

This article originally appeared on Engadget at https://www.engadget.com/googles-bard-ai-chatbot-has-learned-to-talk-070111881.html?src=rss 

Twitter has published an update on how its “Freedom of Speech Not Reach” moderation approach is working, and according to the company, it has seen some encouraging results. In April, the website started limiting the reach of tweets violating its hateful conduct policy and applying a label to them that reads: “Visibility limited: this tweet may violate Twitter’s rules against hateful conduct.” Apparently, Twitter has applied the label to more than 700,000 posts since then and has proactively prevented ads from appearing adjacent to those content. 

The company also said that the label reduces the reach of a post by 81 percent, thereby effectively limiting the visibility of posts that potentially exhibit hateful conduct. In addition, Twitter revealed in its update that more than one-third of users choose to delete labeled tweets themselves once they’ve been notified that they have violated the website’s policy and only four percent of authors have appealed labels. 

The company charging for API access means most researchers studying hate speech can’t independently verify these claims. But Twitter is clearly claiming that its approach has been effective so far. In fact, the website is pushing through with its plan to expand its labels and include more types of policy violations. According to its announcement, it will now also label and downrank posts that violate its Abusive Behavior and Violent Speech policies. Tweets that will be labeled in the coming weeks include posts with malicious content targeting individuals, those that encourage others to harass an individual or group of people, those that threaten to inflict physical harm on others, and tweets that encourage others to commit acts of violence or harm. 

This article originally appeared on Engadget at https://www.engadget.com/twitter-says-it-limited-the-reach-of-over-700000-tweets-that-violated-its-policy-063548086.html?src=rss