Beggers1960

A bunch of English-speaking publications jump on a story about a hypothetical situation in which internet-connected toothbrushes are used to launch a DDoS attack and everyone loses their minds.

Prime example of hype and hysteria on security-related stories.

Mind boggling that folks responsible for evacuating high priority Afghan government workers etc didn't know about basic email privacy practices.

A very public spat emerging in the UK with the cloud competition probe starting to heat up.

AWS and Google Cloud very much knives out at this stage - ironic given the former's significant market share and the fact it's swamped many a smaller provider over the years.

Pretty damning in terms of the gender pay disparity here

Pretty impressive numbers here. Double the capacity of Claude 2.0, and way more than GPT-4. Also cut hallucinations significantly - this'll get enterprise's attention.

Deal finally goes through despite concerns that Chinese regulators might scupper the deal.

"Salesforce CEO Marc Benioff has revealed the firm will offer positions to outgoing OpenAI employees amidst a looming exodus from the beleaguered tech giant."

Benioff cashing in on the revolt at OpenAI in his typical brash and brazen style.

This would be the final, ultimate form of Copilot if I had my way with it.

"The Amazon Olympus model will have two trillion parameters. If correct, this would make it one of the largest and most powerful models currently in production.

"By contrast, OpenAI’s GPT-4, the current market leading model, boasts one trillion parameters."

If this is legit, then Amazon seriously isn't f*cking around. The Titan FMs were already largely outdated by the time of release compared to GPT-4 so they're clearly doubling down on trying to claw back some custom from OpenAI and Microsoft with this move.

If it's released in December or even early next year, it'll be interesting to see how OpenAI reacts to this. Just a perpetual LLM dick swinging contest over the next couple of years.

Find it hard to believe that this is being framed as a 'UK-first' project.

If it is, then I'm genuinely perplexed. This seems to be a fairly frequent tactic on mainland Europe and even across the Irish Sea.

There's a lot to be said about the 'success' of this AI Safety Summit. While there's clearly been some positives out of this, there does seem to be a feeling that it's just highlighted the clear differences in regulatory stances and how some of the larger players, like the US, are going to things moving forward.

Also lol at the US turning up to the UK's big AI event and basically shitting on their plans with the AI Safety Institute announcement.

Momentum growing in terms of global discussions around AI governance and regulation at the moment. Biden administration's executive order, EU AI Act, and the G7 recently outlining responsible use guidelines.

Honestly so disheartening that in 2023 the old bullshit views are still prevalent in an industry that has traditionally framed itself as forward thinking and inclusive.

"...the UK subsidiary was only informed of the incident five minutes before it was announced by the American parent company."

Five minutes warning. Jesus Christ.

Advisory from Microsoft off the back of the recent Gmail anti-spam rule changes. Might cause some headaches.

In fairness that's a very valid point. Certainly have the financial clout to go through with this in terms of talent and the nuts and bolts.

OpenAI going down the ruthless acquisition route with its growth plans. Seems Microsoft certainly has rubbed off on them.

That aside, not an entirely bad idea on their part - actually ensuring this doesn't turn into a money drain, however, is an entirely different conversation. Logistical side of this could also be a shitshow.

“Our market study has identified features that make it more difficult for UK businesses to switch and use multiple cloud suppliers. We are particularly concerned about the position of the market leaders Amazon and Microsoft,” Ofcom said in a statement today.

Microsoft and AWS called out here by Ofcom - 80% market share is madness, but sadly indicative of the current states of the broader global cloud space.

Death, taxes, and industry stakeholders penning open letters criticizing pending EU legislation.

Interesting rundown of the whole FTX calamity and what could be in store for Sam Bankman-Fried

"Apple's chief executive Tim Cook said the firm wants to hire more staff in the UK, in contrast to redundancies seen across the tech sector.

"He said the company wants to take on more staff to work in artificial intelligence (AI)."

Interesting approach from Apple in terms of AI development in recent months. Compared to AWS, Google, Microsoft et al the firm seems to have been keeping a relatively low profile and just innovating quietly without the hype and hyperbole.

"Hundreds of GitHub repositories have been targeted by a threat actor masked as the GitHub platform’s Dependabot feature to install password-stealing malware."

"They have come to the conclusion that there's really no point in maintaining them that long, because people are not using them. So the six-year update policy is going away. When 4.14 goes out of support… early next year… there will not be another six-year kernel to replace it."

Shuttered after just a couple of months - how many organizations went all-in on AR/metaverse over the last 18 months and have unlocked absolutely zero value?

“That means you can get the same work done with fewer people".

Not exactly news that firms view AI as a way to cut headcount, but interesting to see that IBM's chief exec has been so outspoken on the subject in recent months. Already cutting staff in HR and back office roles with the view of automating them long-term.

Sad news. That's both co-founders passed away in the last two years, with Charles Geschke in 2021.

Adobe story with these two is a great example of pioneers being restricted in a corporate environment, going it alone, and making a success of things.

"...unknown to the police, his colleagues, and his employer, Liles commenced a separate and secondary attack against the company.

"Liles accessed board members’ private emails more than 300 times and altered the original ransom note to change the payment address to his own cryptocurrency wallet."

Yikes, talk about creating trust issues. Charged with helping investigate and remediate and then capitalises on the situation to extort their employer.

"The capture of Singapore by Imperial Japanese forces in February 1942 had a monumental impact on the British Empire. To lose such a key strategic position in the Far East at the height of the Second World War was, to put it bluntly, a failure of cataclysmic proportions."

"People in a self-driving car should not be responsible if it crashes, legal watchdogs have proposed.

"The Law Commission of England and Wales, and the Scottish Law Commission, want legal protections to be introduced for owners of self-driving cars if anything goes wrong with the automation.

"This could include occurrences like speeding or running a red light."

Potentially a major spanner in the works for manufacturers here and will likely further delay the introduction of autonomous vehicles on UK roads, Which isn't necessarily a bad thing given notable incidents due to software issues etc.

"Scientists have developed video games using AI technology to diagnose, monitor and treat depression.

"The platform, called Thymia, aims to make depression and other mental health conditions as measurable as physical ailments."

An interesting concept here, especially in regard to measuring / testing mental health issues to potentially improve treatment outcomes.