Think about a world with out Fb. For simply half an hour. A gaggle of AI leaders get collectively however don’t appear to ask Elon Musk. Amazon takes knowledge centres nuclear. A brand new competitor for ChatGPT and Google. And public belief in AI is declining.
All this and extra on the “simply breath, Fb shall be again” version of Hashtag Trending. I’m your host, Jim Love, CIO of IT World Canada and TechNewsDay within the US.
We heard the collective scream as a whole lot of hundreds of customers of Meta’s social media websites – Fb, Instagram and Threads had been locked out of their accounts early this morning. The main points are nonetheless coming in as we file this present for tomorrow morning, however right here’s what we all know.
Between about 10 am Jap time and midday, a lot of customers had been locked out of their accounts and unable to login to Fb, Instagram and Threads.
Preliminary stories posted on Cyberint.com claimed that three menace actor teams – Skynet, Godzilla, and Nameless Sudan – claimed that they’d shut down Fb, Instagram and Threads.
The numbers affected had been within the a whole lot of hundreds. The location Downdetector logged over 560,000 US Fb customers unable to get on the positioning and about 160,000 in Canada. The proportions had been the identical on Instagram and Threads.
However not everybody was affected. Whereas I used to be viewing feedback from customers in Toronto and Ottawa on Downdetector, I used to be capable of login into Fb from my home which is a two hour drive to Toronto.
different websites together with Twitter and others, there was the same spike at 11 am though nothing just like the variety of customers affected on Meta’s websites. These had been bigger than regular however solely within the a whole lot.
I’m certain that there shall be extra on this. Try ITWorldCanada.com for an replace this morning and be careful for extra from Howard Solomon, host of CyberSecurityToday.
I’m certain this shall be a part of the dialogue in the present day and perhaps even in days to come back.
Sources: Downdetector, Cyberint.com, ITWorldCanada.com
As Meta’s customers had been being locked out, Elon Musk took the time to make a transparent assertion about how assaults on any websites are unacceptable. Sorry, simply kidding.
Elon took to X to mock the outage, with a meme and in a separate put up, he additionally added, “if you happen to’re studying this put up, it’s as a result of our servers are working.”
At all times an aesthetic man.
Virtually as if in response to Elon Musk’s lawsuit towards OpenAI, a brand new coalition of AI corporations that has printed an open letter sponsored by SV Angel, a thirty 12 months outdated funding agency headed by founder Ron Conway. Their opening to the letter says, “we put money into and assist founders in AI who’re dedicated to constructing the following technology of AI know-how to maneuver humanity ahead.
The letter says, “We, the undersigned, already are experiencing the advantages from AI, and are dedicated to constructing AI that may contribute to a greater future for humanity – please be a part of us!”
The letter seems to be open to everybody to signal, however the preliminary signatures are a digital “who’s who” in AI – OpenAI, Google, Mistral, Meta, Salesforce, Hugging Face and the checklist goes on and on, with one famous exception. Elon Musk didn’t look like one of many preliminary signatures. Is that as a result of he doesn’t imagine in it?
Or is it that the opposite children don’t need to play with him anymore?
Test it out and signal it at openletter.svangel.com
Sources embrace: svangel.com
Amazon Internet Companies (AWS) has moved to nuclear-powered knowledge facilities to scale back its carbon footprint. A $650 million acquisition cope with Talen Power grants AWS possession of a knowledge heart complicated, located adjoining to Talen’s Susquehanna nuclear energy plant in northeast Pennsylvania.
Initially opened in early 2023 with a 48-megawatt capability, there are plans to develop Cumulus to 475 megawatts, with potential for additional development with Amazon’s funding.
Talen Power will obtain $350 million upon the deal’s closure, with a further $300 million to comply with primarily based on the completion of growth milestones all through 2024. Talen will proceed to produce AWS with direct entry to electrical energy generated by the nuclear energy plant, with AWS anticipating to have as much as 960 megawatts at its disposal within the coming years. If I did my calculation proper, that’s sufficient to energy nearly 1,000,000 houses, simply to place it into perspective.
This is only one of Amazon’s ongoing efforts to energy its knowledge facilities with carbon-free and renewable vitality sources. Final month, Amazon entered an influence buy settlement with a wind farm in Oregon, and it has explored different vitality options, together with pure fuel gas cells, which might be transformed to run on hydrogen.
The transfer into nuclear vitality will not be distinctive to Amazon; the information heart business is more and more contemplating small nuclear reactors (SMRs) for his or her modularity and potential value advantages, regardless of none being operational but.
With the appearance of AI and the rising want for energy for knowledge centres, discovering working options shortly is essential.
Sources embrace:
Anthropic, based by Dario Amodei, former vp of analysis at OpenAI, has been a strong quantity three within the huge generative AI gamers. It initially distinguished itself as a rival to ChatGPT near equal in means, however having a bonus of an enormous context window, allowed it to soak up bigger questions than OpenAI.
Claude.ai, their generative AI mannequin has competed in a approach nearly harking back to the outdated Mac pc in comparison with the larger Home windows machines. It’s been a favorite of writers and those that like it, like it. There’s a devoted fan base on locations like YouTube who sing its praises.
However the latest launch has set YouTube ablaze. Claude 3
now claims that it outperforms OpenAI and different generative AI rivals.
The announcement, made by way of a weblog put up, introduces three foundational fashions named Opus, Sonnet, and Haiku, every designed with enhanced evaluation, forecasting, code technology, and content material creation capabilities.
Claude 3 Opus has trotted out a sequence of scores on assessments that declare to point out superior outcomes such because the HumanEval coding take a look at, scoring 84.9 per cent, in comparison with Google’s Gemini Extremely at 74.4 per cent and OpenAI’s GPT-4 at 67 per cent.
There are a sequence of different measures and they’re all spectacular. Truthful warning, these assessments, like human IQ assessments have been controversial, however nonetheless, for a third-place performer to make a leap like this has made waves.
With Google attempting to launch and Claude.ai making an enormous splash, we will solely marvel if OpenAI will push the event of its subsequent huge transfer. We all know GPT 5 is in testing and rumours abound about its potential, and a brand new growth known as Q*
Sources embrace: CIO Dive, YouTube
Public belief in synthetic intelligence (AI) know-how and the businesses creating it’s declining globally, in response to new knowledge from Edelman and shared with Axios.
This development is obvious each in america and internationally, with belief ranges dropping from 61 per cent to 53 per cent globally over the previous 5 years. Within the U.S., belief has seen a major lower of 15 proportion factors, from 50 per cent to 35 per cent in the identical interval.
Edelman’s international know-how chair, Justin Westcott, emphasizes the significance of belief within the AI period and requires corporations to handle the broader implications of AI know-how, together with its social affect and the safety of private privateness.
The research additionally highlights a political divide in belief ranges inside the U.S., with Democrats displaying barely larger belief in AI corporations (38 per cent) in comparison with independents (25 per cent) and Republicans (24 per cent). Moreover, the know-how sector, which was as soon as essentially the most trusted business in 90 per cent of the nations Edelman research, now holds the highest spot in solely half of these nations.
Curiously, the acceptance of AI varies considerably between developed and creating nations. In developed nations reminiscent of France, Canada, Eire, the UK, the U.S., Germany, Australia, the Netherlands, and Sweden, there’s a robust resistance to the rising use of AI. In distinction, creating markets like Saudi Arabia, India, China, Kenya, Nigeria, and Thailand present a a lot larger acceptance charge.
Edelman’s analysis additionally factors to public dissatisfaction with authorities efforts to manage AI, with many feeling that present laws don’t meet their expectations. This sentiment underscores the pressing want for regulators to handle public considerations about AI extra successfully.
Sources embrace: Axios
That’s our present for in the present day.
Love your feedback.
Ship us a word at [email protected] or drop us a remark underneath the present notes at itworldcanada.com/podcasts – search for Hashtag Trending.
Thanks for listening and have a Fantastic Wednesday.