DIRS.INFO The most relevant business & tech news

☆ How Meta could benefit from the OpenAI shakeup
▼ + stars: | 2023-11-22 | by ( Jonathan Vanian | In | ) www.cnbc.com time to read: +3 min

It could help Meta boost its open-source Llama AI initiatives, as some companies look to diversify away from relying on a single company's large language model. Its AI research team is considered, with Alphabet's DeepMind, one of the most esteemed groups in the tech industry. Yann LeCun, Meta's AI chief, responded to the post with a curt "Yup." Meta could benefit if companies continue to seek multiple AI vendors, much like firms now rely on multiple cloud providers. The more developers access and improve Llama, the more Meta can potentially lower its overall operating and technology research costs, among other benefits.

Persons: Mark Zuckerberg, Sam Altman's, There's, Satya Nadella, Altman, Nadella, Zuckerberg, Meta, Alphabet's DeepMind, MSFT, Yann LeCun, curt, Yup, Sam Altman Organizations: Microsoft, Meta, Technologists, Twitter, CNBC

June 27 (Reuters) - Reka, an AI model startup founded by former researchers from Alphabet's DeepMind (GOOGL.O), has raised $50 million in a fresh round of funding from investors including Yuri Milner's DST Global Partners and Snowflake (SNOW.N), the companies told Reuters. Reka focuses on building large and specialized AI models for enterprise customers. Snowflake, which invested in the company, also announced a partnership to allow users to run third-party model providers like Reka within their Snowflake account. The partnerships came as Snowflake competitor Databricks acquired AI model training company MosacicML in a $1.3 billion deal on Monday. Yogatama said Reka has built both large models and proprietary model distillation technology to personalize models for specific use cases.

Persons: Alphabet's, Yuri Milner's, Christian Kleinerman, Databricks, Dani Yogatama, Yogatama, Rob Toews, Krystal Hu, Mark Potter, Anna Driver Organizations: Global Partners, Reuters, Nvidia, Radical Ventures, Thomson Locations: Snowflake, Reka, Toronto

June 27 (Reuters) - Reka, an AI model startup founded by former researchers from Alphabet's DeepMind (GOOGL.O), has raised $50 million in a fresh round of funding from investors including Yuri Milner's DST Global and Snowflake (SNOW.N), the companies told Reuters. Reka focuses on building large and specialized AI models for enterprise customers. Snowflake, which invested in the company, also announced a partnership to allow users to run third-party model providers like Reka within their Snowflake account. The partnerships came as Snowflake competitor Databricks acquired AI model training company MosacicML in a $1.3 billion deal on Monday. Yogatama said Reka differentiated by building both large models and proprietary model distillation technology to downsize models for specific use cases.

Persons: Alphabet's, Yuri Milner's, Christian Kleinerman, Databricks, Dani Yogatama, Yogatama, Rob Toews, Krystal Hu, Mark Potter Organizations: Yuri Milner's DST, Reuters, Nvidia, Radical Ventures, Thomson Locations: Reka, Toronto

☆ Elon Musk cut off OpenAI's access to Twitter's data because he felt the company wasn't paying enough, report says
▼ + stars: | 2023-04-28 | by ( Beatrice Nolan | ) www.businessinsider.com time to read: +2 min

In December, Elon Musk shut off OpenAI's access to Twitter's data, the New York Times reported. Musk reportedly felt the AI company was not paying enough for access, per the publication. In December, only weeks after OpenAI launched ChatGPT, the company had its access to Twitter's data shut down, per a report. Musk felt that the approximately $2 million a year OpenAI was paying to license the company's data was not enough,The AI company had been licensing Twitter's data to help build its AI chatbot, per the NYT. Representatives for OpenAI and Elon Musk did not immediately respond to Insider's request for comment, made outside normal working hours.

☆ Elon Musk says people should be cautious with AI because the tech is 'a danger to the public'
▼ + stars: | 2023-04-19 | by ( Beatrice Nolan | ) www.businessinsider.com time to read: +2 min

Elon Musk called AI a danger to the public during an interview with Tucker Carlson. In an interview on Fox News' "Tucker Carlson Tonight," he said people should be "cautious with AI" and advised that there should be government oversight because the tech is "a danger to the public." "I think we'll have a better chance of advanced AI being beneficial to humanity in that circumstance," he added. It's far from the first time the billionaire has discussed the potential risks of advanced AI. He's previously compared AI with nuclear warheads, told Tesla investors that the tech stresses him out, and criticized OpenAI, a company he cofounded.

☆ DeepMind's CEO said there's a chance that AI could become self-aware in the future
▼ + stars: | 2023-04-17 | by ( Sawdah Bhaimiya | ) www.businessinsider.com time to read: +3 min

The CEO of Alphabet's DeepMind said there's a possibility that AI could become self-aware one day. DeepMind is an AI research lab that was co-founded in 2010 by Demis Hassabis. The CEO of Alphabet-owned AI research lab, DeepMind Technologies, spoke about the potential of artificial intelligence in an interview with CBS' "60 Minutes," which aired on Sunday. DeepMind CEO Demis Hassabis told CBS that he thinks that AI might one day become self-aware. Hassabis told CBS that he believes AI is "the most important invention that humanity will ever make."

☆ Elon Musk is moving forward with a new generative AI project at Twitter after purchasing thousands of GPUs
▼ + stars: | 2023-04-11 | by ( Kali Hays | ) www.businessinsider.com time to read: +4 min

Tech companies typically use GPUs to work on large AI models, given the computational workload the newer technology requires. Musk, however, has criticized the recent development of generative AI, saying the technology is powerful and needs regulation to make sure it's operating within "the public interest." It's unclear exactly what Twitter will use generative AI for, the people familiar said. Generative AI has the capability, if trained for the use, to create new advertising images and text to target specific audiences. Nvidia, which is estimated to have 95% of the market, manufactures a GPU for large AI models that costs $10,000.

☆ Tech leaders are urging caution on AI
▼ + stars: | 2023-03-30 | by ( Paayal Zaveri | ) www.businessinsider.com time to read: +4 min

Insider asked ChatGPT, the viral AI chatbot sweeping the internet, to whip up a layoff memo for a pretend tech company, Gomezon. Elon Musk, Steve Wozniak, researchers at Alphabet's DeepMind, and other AI leaders are calling for a pause on training AI models more powerful than OpenAI's GPT-4. My colleague Emilia David looked at why Elon Musk and other tech leaders are right: AI needs to slow down. An Apple Watch is an essential for many of us these days, but the right band can make all the difference. Check out Insider's review of the 18 best Apple Watch bands in 2023.

☆ How an AI breakthrough from Alphabet's DeepMind 'changed scientific history' and could lead to new cures
▼ + stars: | 2023-02-28 | by ( Andrew Dunn | ) www.businessinsider.com time to read: +11 min

"AlphaFold has sparked a wave of innovation by showing people what's possible," said Chris Bahl, the chief scientist at AI Proteins, a Boston startup using AlphaFold to help develop drugs. "AlphaFold, amazing as it, is just the beginning," Demis Hassabis, the CEO of DeepMind, said on a podcast last year. AlphaFold2 was built with far more biological and physics knowledge of proteins, Jumper said. Next uses will be 'progressively harder' as DeepMind stays secretive on its future workJohn Jumper, a senior staff research scientist at DeepMind who helped develop AlphaFold. "But AI will also continue to progress rapidly, and the folks at DeepMind are very good, so I'm optimistic."

☆ ChatGPT could hypothetically get hired as an entry level coder if it interviewed at Google, internal document reportedly says
▼ + stars: | 2023-02-01 | by ( Sawdah Bhaimiya | ) www.businessinsider.com time to read: +2 min

ChatGPT could get hired as an entry-level software engineer at Google, internal communications cited by CNBC said. "Amazingly ChatGPT gets hired at L3 when interviewed for a coding position," a note reported by CNBC said. "Amazingly ChatGPT gets hired at L3 when interviewed for a coding position," a note in one internal document comparing LaMDA and ChatGPT said, per CNBC. When asked whether ChatGPT and AlphaCode — an AI coding system designed by Alphabet's DeepMind — would replace programmers, ChatGPT and LaMDA both disagreed, per CNBC. At a December all-hands meeting, CNBC reported, Google staff raised concerns that the company may be falling behind in the AI space.

☆ How Microsoft can double its ad business to $20 billion
▼ + stars: | 2022-11-28 | by ( Jordan Parker Erb | ) www.businessinsider.com time to read: +5 min

As the ad industry braces itself for what's sure to be a challenging 2023, we're taking a look at what Microsoft will need to do to achieve its ambitious goal of doubling its advertising business to $20 billion. Microsoft plans to grow its advertising business to $20 billion. In an interview with Insider, Microsoft Ads chief Rob Wilk shared plans to double the size of the company's ad business. Were Microsoft to reach $20 billion in ad revenue, it would overtake Chinese tech and media giant Tencent to become the sixth-largest digital ad seller worldwide, based on Insider Intelligence's estimates. Gyasi Calhoun, a front-end software engineer and developer at Twilio, said there's perks to the job, but that the industry can be stressful.

☆ Alphabet's DeepMind has put a hiring freeze on new interns, canceling one applicant's interview with just an hour's notice
▼ + stars: | 2022-11-27 | by ( Jyoti Mann | ) www.businessinsider.com time to read: +3 min

DeepMind said it is pausing the hiring of new interns while it reviews its "strategic priorities." Another said she is "gutted" that it froze hiring and said the "tech hiring freeze is hitting hard." The candidate wrote: "Disappointed & shocked to hear that DeepMind 'paused' intern hiring and canceled all scheduled interviews, just one hour before my interview ... Another aspiring DeepMind intern tweeted that she was "gutted" to hear about the freeze as she had made it to the final round of the hiring process. "The tech hiring freeze is hitting hard," she added.

☆ Amazon's moonshots lab recently shut down several secret projects, shocking members of the team. Here are the initiatives that got the chop, and the ones that survived.
▼ + stars: | 2022-10-25 | by ( Eugene Kim | ) www.businessinsider.com time to read: +4 min

Amazon's Grand Challenge division recently shut down three confidential projects, a cost-cutting move that shocked and disappointed members of the secretive moonshots lab, Insider has learned. Two healthcare projects survived the cull. The healthcare projects that survived are a new type of at-home fertility-monitoring device and a therapeutic treatment to protect against antibiotic resistance, these people said. The decision to keep two healthcare projects reflects CEO Andy Jassy's ambition in this field. Glow, a video-calling device that Amazon shut down earlier this month, was also part of the broader Grand Challenge team.

Search resuls for: "Alphabet's DeepMind"

13 mentions found