{"id":594872,"date":"2026-04-09T04:19:13","date_gmt":"2026-04-09T04:19:13","guid":{"rendered":"https:\/\/www.newsbeep.com\/au\/594872\/"},"modified":"2026-04-09T04:19:13","modified_gmt":"2026-04-09T04:19:13","slug":"mark-zuckerberg-mounts-meta-ai-as-anthropics-claude-mythos-preview-pushes-them-ahead-of-the-pack","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/au\/594872\/","title":{"rendered":"Mark Zuckerberg mounts Meta AI as Anthropic\u2019s Claude Mythos Preview pushes them ahead of the pack"},"content":{"rendered":"

\"David<\/p>\n

April 9, 2026 \u2014 11:48am<\/p>\n

Save<\/p>\n

You have reached your maximum number of saved items.<\/p>\n

Remove items from your saved list<\/a> to add more.<\/p>\n

Save this article for later<\/p>\n

Add articles to your saved list and come back to them anytime.<\/p>\n

Got it<\/p>\n

AAA<\/p>\n

Mark Zuckerberg spent the better part of a billion dollars \u2013 some reports suggest considerably more – wooing AI researchers to dinner at his Lake Tahoe compound over the past year. He handed out nine-figure pay packages – in some cases to Australians<\/a> \u2013 and effectively blew up his entire AI operation. He fired people, hired new people, and installed a 29-year-old as the man responsible for salvaging his company\u2019s reputation in the field he has declared the defining bet of his career.<\/p>\n

On Thursday, the first results arrived.<\/p>\n

Muse Spark<\/a>, Meta\u2019s new AI model, landed with enough fanfare to push the company\u2019s stock up 6.5 per cent in a single day. It\u2019s the first step in what Zuckerberg has promised investors will be \u201cpersonal superintelligence for everyone.\u201d<\/p>\n

\"MetaMeta chief executive Mark Zuckerberg made wholesale changes to the company\u2019s AI operation.AP<\/p>\n

The model itself is good, by all accounts, in what\u2019s a fiercely competitive race with incredibly high stakes. Independent testers who got early access say Meta is now, credibly, a competitive AI lab for the first time in years. The model performs well on reasoning benchmarks, handles health queries with depth and draws on social content \u2013 think Instagram posts, Facebook threads, and Reels – that no competitor can replicate.<\/p>\n

That last piece is potentially very valuable and is underpinning Meta\u2019s AI competitive strategy. It is also, depending on your view of the tech giant\u2019s relationship with privacy and personal data, something worth watching carefully.<\/p>\n

\"MuseMuse Spark, Meta\u2019s new AI model, can help with nutritional information.<\/p>\n

What it isn\u2019t, and what Meta has acknowledged directly, is cutting-edge across the board. Coding remains a weak spot while long-horizon agentic tasks \u2013 the kind where an AI works autonomously through complex, multi-step problems \u2013 are still a work in progress.<\/p>\n

To understand why Wednesday\u2019s announcement mattered so much you have to go back just over a year, to when Meta\u2019s previous model Llama 4<\/a> arrived with a thud.<\/p>\n

It underperformed, and was followed by revelations that Meta had been quietly gaming a third-party benchmark used by the industry to rank models \u2013 essentially, fiddling the scoreboard. The company later admitted as much. Then, the tech giant\u2019s biggest planned model, something called Behemoth, was shelved without ever being released publicly<\/a>. Senior AI staff left, with some heading straight to competitors.<\/p>\n

Related Article\"Canva<\/a><\/p>\n

It was, by any reading, an embarrassing period for a company that had loudly proclaimed its open-source AI strategy as a gift to humanity and a competitive masterstroke simultaneously.<\/p>\n

Zuckerberg, as he usually does, didn\u2019t respond with humility but instead with money, and quick reorganisation. He brought in Alexandr Wang, the former Scale AI chief executive whose data-labelling startup Meta had just been valued at $US14 billion, and handed him the keys to a new entity called Meta Superintelligence Labs. Wang assembled a tight inner circle of elite researchers and set them to work.<\/p>\n

\"AlexandrAlexandr Wang, who co-founded Scale AI with Lucy Guo.The New York Times<\/p>\n

Their first publicly released work \u2013 Muse Spark \u2013 is what came out today. It\u2019s a closed model, meaning that Meta holds on to the underlying code, training data and model weights. Meta says it hopes to open-source future versions.<\/p>\n

Meanwhile, across town in San Francisco just a couple of days earlier, Anthropic set off a different kind of alarm.<\/p>\n

\"AnthropicAnthropic chief executive Dario Amodei. The AI giant seeks to embed itself as the general-purpose artificial intellgience platform of choice for the business world.Bloomberg<\/p>\n

Where Meta was managing investor expectations and talking up health chatbot features, Anthropic was doing something unusual: announcing a model it says is too dangerous to release to the public<\/a>.<\/p>\n

Claude Mythos Preview<\/a>, the company disclosed this week, is already capable of finding and exploiting so-called zero-day vulnerabilities in software \u2013 flaws that even the software\u2019s own developers don\u2019t know exist.<\/p>\n

Those capabilities weren\u2019t merely hypothetical, either \u2013 the model has reportedly already located a 27-year-old bug in OpenBSD, the operating system quietly underpinning a significant chunk of the world\u2019s secure network infrastructure. It also found a 16-year-old flaw in video encoding tool FFmpeg and has catalogued vulnerabilities in Linux software running on most of the world\u2019s servers.<\/p>\n

Related Article\"OpenAI\u2019s<\/a><\/p>\n

Rather than a public launch, Anthropic is routing access through something it calls Project Glasswing \u2013 a coalition of more than 40 companies, including Apple, Amazon, Microsoft, Google, CrowdStrike and Palo Alto Networks, backed by $US100 million in usage credits. The idea is to use AI to find and patch vulnerabilities before malicious actors do the same thing.<\/p>\n

The juxtaposition of these two stories in the same week is something worth sitting with.<\/p>\n

Meta is celebrating because it has rebuilt itself into a credible AI competitor after a humiliating collapse. While Anthropic \u2013 which announced its own annual revenue had tripled to more than $US30 billion \u2013 is simultaneously ringing the loudest alarm bell the industry has heard in some time, warning that AI\u2019s capabilities have now outpaced the security protecting some of the world\u2019s critical infrastructure.<\/p>\n

Zuckerberg\u2019s model can recommend nearby restaurants while Anthropic\u2019s model, which the company has decided you cannot be trusted with yet, can dismantle security systems that have been considered safe for decades.<\/p>\n

Both things are true at the same time, and sum up what you need to know about the current state of AI in April 2026.<\/p>\n

The Business Briefing newsletter delivers major stories, exclusive coverage and expert opinion. Sign up to get it every weekday morning<\/a>.<\/p>\n

Save<\/p>\n

You have reached your maximum number of saved items.<\/p>\n

Remove items from your saved list<\/a> to add more.<\/p>\n

\"DavidDavid Swan<\/a> is the technology editor for The Age and The Sydney Morning Herald. He was previously technology editor for The Australian newspaper.Connect via X<\/a> or email<\/a>.From our partners<\/p>\n","protected":false},"excerpt":{"rendered":"April 9, 2026 \u2014 11:48am Save You have reached your maximum number of saved items. Remove items from…\n","protected":false},"author":2,"featured_media":594873,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[256,254,255,64,63,105],"class_list":{"0":"post-594872","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-au","12":"tag-australia","13":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/594872","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/comments?post=594872"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/posts\/594872\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media\/594873"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/media?parent=594872"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/categories?post=594872"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/au\/wp-json\/wp\/v2\/tags?post=594872"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}