Anthropic unveils the latest Claudes with claim to AI coding crown
Why it matters: Competition is hot between Anthropic, Google and OpenAI for the "best frontier model" crown as questions persist about the companies' ability to push current AI techniques to new heights.
Driving the news: At the high end, Anthropic announced Claude 4 Opus, its "powerful, large model for complex challenges," which it says can perform thousands of steps over hours of work without losing focus.
Claude Sonnet 4 — Anthropic's "smart, efficient model for everyday use" — is designed to replace Claude Sonnet 3.7 with improved coding abilities and better adherence to instructions.
Both are hybrid models, meaning they can answer users straightaway or take additional compute time to perform reasoning steps.
In addition to the new models, Anthropic announced a series of other new capabilities for developers who connect to its models, including the ability to combine reasoning with the use of the Web and other tools.
What they're saying: "AI agents powered by Opus 4 and Sonnet 4 can analyze thousands of data sources, execute long-running tasks, write human-quality content, and perform complex actions," Anthropic said in a statement.
Between the lines: Anthropic is making one change in its reasoning mechanics — it will now aim to show summaries of the models' thought processes rather than trying to document each step.
The big picture: The announcements, made at Anthropic's first-ever developer conference, come after a busy week in AI that saw Microsoft announce a new coding agent and a partnership to host Elon Musk's Grok, Google expand its AI-powered search efforts and OpenAI announce a $6.5 billion deal to buy io, Jony Ive's secretive AI hardware startup.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
CNET
32 minutes ago
- CNET
I Tried AI for Finding Cheap Flights, but It Was a Bumpy Ride
If you're planning s summer vacation, you've probably thought about using an artificial intelligence chatbot to help save money on flights. Hunting for flights is all part of the summer travel adventure, but it can quickly take a turn for the worse when I feel like I'm playing cat and mouse with Google's algorithm. Log out. Turn the VPN on. Use an incognito browser window. Switch between my MacBook and iPhone. These days, any strategies often fall short when you try to "trick" the algorithm so that it doesn't hike up the cost of the flight after seeing your repeated interest in that route and those dates. Google Flights is my favorite booking platform, so I wondered if I could use Google's AI, Gemini, be the new most underrated travel hack for 2025? With the price of flights these days, I sure hoped so. Gemini was released in March 2023 (when it was known, briefly, as Bard) and is one of the biggest AI models on the market. It's free to use if you have a Google or Workspace account with admin-enabled access to Gemini. There's also a paid version, which is $20 a month, with a two-month free trial. See also: Traveler's Prep Checklist: Tips for Easier Flights and Road-Trips For me, an Australian who lives in the US, the cost of flying home and back is in the thousands. Any savings AI can find would make a world of difference to my bank account. How to use AI to find cheap flights With a dedicated Google Flights extension, I had expected Gemini to transport me to the best deals. I followed the prompts to enable extensions, but I couldn't get it to work -- even after poking around in my Google admin console all afternoon. A CNET editor was able to use this tool as expected (see screenshot below), so it's worth testing out to see if it works for you. Gemini / Screenshot by CNET Annoyed by the lack of ease, I ran a quick search for AI tools that could save me money on plane tickets, and in my "travels" I discovered Karma, an AI coupon code extension that I could easily install. I saw Expedia on its list of affiliated brands, so I used that travel platform to search for flights. I picked dates four months away for a hypothetical trip home to Australia. The cheapest return flight was $1,443. Not bad, but Karma pinged me with a coupon. I clicked "apply coupons." See also: Worried About Expensive Holiday Airfare? 5 Easy Tips to Save Money on Plane Tickets I copied the code and continued in Expedia, assuming it would automatically apply at the checkout page, but there was no area to apply a code. Turns out Karma coupons can't be applied to all Expedia purchases; maybe I could have found a way to use one somehow, but it wasn't working for this flight option. But for me, it's strike number two. Back to Gemini, but for some advice this time. I asked Gemini how AI can help me save money on flights. The answers were to be expected, but two words piqued my interest: "secret fares." Gemini / Screenshot by CNET I checked the age of those two links Gemini referenced and they were from 2018, so I responded to Gemini asking what platforms have "secret fares." Gemini / Screenshot by CNET While skiplagged sounds like an adventure in itself, I'm too old to fly with no check-in luggage or risk being blacklisted by airlines. When there's only one (practical) way to enter my country (by air), the potential reward wasn't worth the risk. It was interesting that Hopper wasn't mentioned in Gemini's list, although it was the source of the original "secret fares" information. It had come up in my own research, so I decided to give it a shot. Hopper's Price Prediction function is only available in the app. I downloaded it. See also: Travel Planning With AI: I Tested It for a City I Know Inside and Out Once I set myself up in Hopper, I ran a quick Google Flights search and found a good option for $1,256. Hopper's cheapest option was $1,272 (for that same flight), but it said I could save up to $199 by waiting longer before booking. I turned on Price Watch notifications. And waited. I checked back in a day later and it hadn't changed, then the following day it had gone up. I opened the app a few days later and again it had gone up, not down. If you have more time, it could take a few weeks for Hopper to identify a lower price. As always, there's no guarantee. Strike number three. Double-check flight pricing If you're using a flight booking platform like Google Flights, Skyskanner or Expedia, these sites do have AI features built in. This might be sufficient, and you can use AI chatbots on those sites for travel advice. For confirmation, I checked with ChatGPT. It gave me tools I'd already tried out. ChatGPT/Screenshot by CNET What I did discover via ChatGPT is the new Google Flights refund feature on itineraries departing from the US. If the price drops on a booking (that had a price guarantee badge), you get refunded through the Google Pay app. Pretty cool. That one feature alone is enough for me to stick with Google Flights, alongside accessing advice from AI chatbots. You can also use AI to plan out a strategy to use credit card points to save on flights. However, AI has been found to pull up outdated card bonuses and rewards information, so it's important to vet it before applying for a new credit card. The verdict on finding cheap flights using AI After multiple attempts to use AI to save money on flights, I unpacked my bags. While I expected Gemini's Google Flights extension to be as smooth as Google Flights itself, I couldn't get it to work, even with the features enabled in my Google account. Gemini also admits it doesn't always get it right, so be sure to check the flight details on the airline's website. Given flight tickets are often in the thousands, you don't want to risk any mistakes. In short, using AI to find flight deals still has a way to go. I did, however, find some interesting things via AI, such as the refund feature. AI is a great thought partner but not yet a booking partner.
Forbes
33 minutes ago
- Forbes
‘Please Be Careful'—ChatGPT's OpenAI Issues Serious Warning
ChatGPT-maker OpenAI has become one of the world's most valuable private companies, scoring a $300 billion valuation earlier this year. Front-run Donald Trump, the White House and Wall Street by subscribing now to Forbes' CryptoAsset & Blockchain Advisor where you can "uncover blockchain blockbusters poised for 1,000% plus gains!" OpenAI's massively popular ChatGPT chatbot helped it hit $10 billion in annual recurring revenue last month, with the runaway success of ChatGPT driving huge investor interest in the privately held company. Now, as the likes of trading app Robinhood use blockchain technology to offer people tokenized exposure to private companies including OpenAI and Elon Musk's SpaceX, OpenAI has warned people to 'be careful' of the tokens. Sign up now for the free CryptoCodex—A daily five-minute newsletter for traders, investors and the crypto-curious that will get you up to date and keep you ahead of the bitcoin and crypto market bull run OpenAI, led by chief executive Sam Altman, has warned people to "be careful" due to Robinhood's ... More "OpenAI tokens." 'These 'OpenAI tokens' are not OpenAI equity,' an OpenAI official X account posted. 'We did not partner with Robinhood, were not involved in this, and do not endorse it. Any transfer of OpenAI equity requires our approval—we did not approve any transfer. Please be careful.''These tokens give retail investors indirect exposure to private markets, opening up access, and are enabled by Robinhood's ownership stake in a special purpose vehicle,' a Robinhood spokesperson said in response to the OpenAI post.
CNET
33 minutes ago
- CNET
Pay Up, AI Bot: That's the Message From a Key Company in How the Internet Works
AI companies might find it harder to access the entire web to train their large language models after the internet infrastructure provider Cloudflare said this week it would block AI data crawlers by default. It's the latest front to open in an ongoing fight between the creators of content and the AI developers who use that content to train generative AI models. In court, authors and content creators are suing major AI companies for compensation, saying copyrighted content was used without permission. (Disclosure: Ziff Davis, CNET's parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.) While content providers are seeking compensation for information that was used to train models in the past, Cloudflare's move marks a new defensive measure against future efforts to train models. But it isn't just about blocking crawlers: Cloudflare says it wants to create a marketplace where AI companies can pay to crawl and scrape a site, meaning the provider of that information gets paid, and the AI developer gets permission. "That content is the fuel that powers AI engines, and so it's only fair that content creators are compensated directly for it," Cloudflare CEO Matthew Prince said in a blog post. Why websites want to block AI crawlers Crawlers -- bots that visit and copy the information from a website -- are a vital component of the connected internet. It's how search engines like Google know what's on different websites, and how they can serve you the latest information from places like CNET. AI crawlers pose distinct challenges for websites. For one, they can be aggressive, generating unsustainable levels of traffic for smaller sites. They also offer little reward for their scraping: If Google crawls a site for search engine results, it will likely send traffic back to that site by including it in search results. Being crawled for training data might mean no additional traffic or even less, if people stop visiting the site and rely just on the AI model. Read more: AI Essentials: 29 Ways to Make Gen AI Work for You, According to Our Experts That's why executives from major websites like Pinterest, Reddit and several major publishing companies (including Ziff Davis, which owns CNET) cheered Cloudflare's news in statements. "The whole ecosystem of creators, platforms, web users and crawlers will be better when crawling is more transparent and controlled, and Cloudflare's efforts are a step in the right direction for everyone," Reddit CEO Steve Huffman said in a statement. Asked about Cloudflare's announcement, OpenAI said its ChatGPT model is intended to help connect its users to content on the web, similarly to search engines, and that it has integrated search into its chat functions. The company also said it uses a separate model from what Cloudflare has proposed to allow publishers to indicate how AI crawlers should behave, known as OpenAI said the model works already and Cloudflare's changes are unnecessary. The training data tug-of-war AI models require a ton of data to train. That's how they're able to provide detailed answers to questions and do a decent (if imperfect) job of providing a wide range of information. These models are fed incredible amounts of information and make connections between words and concepts based on what they see in that training data. The issue is how developers have gotten that data. There are now dozens of lawsuits between content creators and AI companies. Two saw major rulings just last week. In one case, a federal judge ruled Anthropic followed the law when it used copyright-protected books to train its model Claude -- via a concept called fair use. At the same time, the judge said the company's creation of a permanent library of the books was not, and ordered a new trial on those piracy allegations. In a separate case, a judge ruled in favor of Meta in a dispute between the company and a group of 13 authors. But Judge Vince Chhabria said the ruling in this case doesn't mean future cases against Meta or other AI companies will go the same way, essentially that "these plaintiffs made the wrong arguments and failed to develop a record in support of the right one." The idea of charging crawlers to visit a site isn't entirely new. Other companies, like Tollbit, offer services that allow website owners to charge AI companies for crawling. Will Allen, head of AI control, privacy and media products at Tollbit, said the environment around this technology is still developing. "We think it's very early for a content marketplace to form, and we are just starting to experiment here," he told CNET. "We're excited to see many different models flourish." CNET's Imad Khan contributed to this report.
