Site icon IT World Canada

Hashtag Trending Jan.31- Ransomware payments declining? ChatGPT leaking chat histories; UK law may ban Apple from issuing security updates worldwide

Hashtag Trending Podcast

 

Hey out there. Someone wrote me yesterday to tell me that they had tried to pass on how to subscribe, but it was difficult to find Hashtag Trending – apparently a lot of similar names. 

Changing the name of the podcast is a big deal, but finding us isn’t. Just go to our page at itworldcanada.com/podcasts.  Look in any of the daily scripts and there are links that take you directly to us on Apple, Google or Spotify.  

Feel free to share a link to any of those pages with your friends who might want to subscribe. One click and they are there. And I hope I’ve said it enough – thank you for recommending us to your friends and colleagues. It’s working.

Ransomware payments may be on the decline. ChatGPT may be leaking information and no-one is quite sure why, Microsoft’s Future of work claims some huge gains in productivity – and some issues with accuracy. And…they are called Robot wranglers and they’re managing mischievous machines. 

All this and more this edition of, hey, robots have rights too version of Hashtag Trending. I’m your host Jim Love, CIO of IT World Canada and TechNewsDay in the US.  

According to a new report from ransomware negotiation firm Coveware, there has been a significant decline in the number of ransomware victims who choose to pay hackers. In the last quarter of 2024, only 29 per cent of organizations opted to pay a ransom to retrieve their stolen data and unlock their systems during a cyberattack. This figure marks a stark contrast to the first quarter of 2019, when 85 per cent of affected organizations were paying ransoms.

The average ransom payment in the fourth quarter of 2023 was approximately $568,000, which is a 33 per cent decrease from the third quarter. Coveware attributes this decline to several factors. Enterprises have strengthened their cyber defenses and have more robust data backups, enabling quicker recovery from attacks. Additionally, there’s a growing distrust among companies towards hackers’ promises to delete stolen data after receiving the ransom.

Ransomware has emerged as a top cyber threat for organizations of all sizes over the past five years. Government officials have been actively working to reduce the number of ransomware attacks targeting businesses, governments, and other entities. However, ransomware hackers are known for their adaptability and may simply change their tactics, so let’s not count them out just yet.

Sources include: Axios

A user of ChatGPT, Chase Whiteside, reported an alarming incident where he found chat histories from unrelated users in his account. These chats contained sensitive information, including unpublished research papers and other private data. OpenAI officials have stated that this issue resulted from Whiteside’s ChatGPT account being compromised, with unauthorized logins traced back to Sri Lanka. Whiteside, who logs in from Brooklyn, New York, expressed doubts about his account being compromised, citing the use of a strong, unique password.

OpenAI’s investigation suggests that the incident was not a case of ChatGPT leaking chat histories to unrelated users but rather a consequence of an account takeover. This situation highlights the lack of mechanisms such as two-factor authentication (2FA) or the ability to track IP locations of logins on the ChatGPT platform, which are standard security features on most major platforms.

The original suspicion that ChatGPT was leaking private conversations, including login credentials and personal details of unrelated users, has been refuted by OpenAI’s findings. However, the incident underscores the importance of maintaining robust security measures for online accounts and being cautious about sharing personal details in AI service queries.

This incident, along with similar experiences reported in the past, serves as a reminder of the potential risks associated with using AI services and the importance of safeguarding personal and proprietary data.

Sources include: ArsTechnica

Proposed amendments to the UK’s Investigatory Powers Act (IPA) could potentially ban Apple from issuing security updates worldwide, a move that Apple has labeled as an “unprecedented overreach.” This development raises significant concerns about data security and information privacy, not just for British citizens, but for tech users globally.

The IPA, enacted in 2016, already grants the UK government the power to issue orders to tech companies to build backdoors into their products to break encryption. Apple has been a vocal opponent of this, previously threatening to withdraw iMessage and FaceTime from the UK market rather than compromise end-to-end encryption.

The new amendments to the IPA could allow the UK government to prevent Apple from providing security updates to iOS if they interfere with the operations of UK security services. Apple argues that this could force non-UK companies to undermine the security of all users because of their UK user base. The company also points out that the Home Office proposes that the extraterritorial scope of the IPA should apply globally, regardless of a provider’s physical presence in the UK.

The proposed amendments have now moved to the next stage in the UK’s legislative process, first passing through the House of Commons and then to the House of Lords. The Lords, an unelected body, often provide a more considered response to legislation, potentially influencing the final outcome of these controversial measures.

Apple, backed by various civil liberties groups, has strongly objected to these new powers. They argue that the amendments would transform private companies into extensions of the surveillance state and erode the security of devices and the internet globally.

This situation highlights a significant conflict between government surveillance desires and the privacy and security of global tech users. The outcome of this legislative process could have far-reaching implications for tech companies and users worldwide.

I hear a Tom Petty song here – won’t back down. I’m not sure Apple can.

Sources include: 9TO5 Mac

Microsoft’s recent report titled “New Future of Work” acknowledges that AI can speed up certain tasks, like writing, by 37 per cent, but also notes downsides – a 19 per cent decrease in accuracy in work done by experts from Boston Consulting Group using large language models.

Microsoft claims its Copilot is “mostly neutral” in its effects on quality and touts it as a solution to AI-related problems. A survey of enterprise users with access to Copilot showed that 68 percent believed it improved the quality of their work, although this is based on perception.

However, since the announcement of Copilot Pro, a premium version of the OpenAI-powered LLM bot, there have been complaints about its effectiveness, pricing, and even why it exists. 

The report also mentions that LLMs are most helpful to less experienced workers.  

It also reveals that small semantic differences in writing prompts can lead to vastly different results. That’s why Microsoft’s Copilot Lab  is providing a collection of suggested prompts.

Microsoft’s vision of the future of work, heavily influenced by its AI investment, is facing regulatory scrutiny. The US Federal Trade Commission, the European Commission, and the UK Competition and Markets Authority are investigating whether the Microsoft/OpenAI partnership could lead to an anticompetitive situation.

This EU regulation thing isn’t going away, is it?

Sources include: YouTube and The Register

At a subsidiary of GE Appliances in LaFayette, Georgia, Scott Samples is pioneering a new profession: he’s a ‘robot wrangler.’ This new job has emerged as companies increasingly integrate robots into their operations, only to discover that these automatons sometimes need a bit of human guidance.

One notorious robot, nicknamed Blinky for its flashing lights signaling distress, is known for its mischievous antics. It often disrupts the smooth transfer of items due to its overzealous conveyor belt movements, leading to humorous exasperation among the staff. “Oh s—, not again. It’s him again,” they’d say, attributing a personality to the mechanical troublemaker. Blinky, unsurprisingly, had no comment on these allegations.

Scott Samples, the ‘robot wrangler’ at Roper, a GE Appliances subsidiary, is the go-to guy for robots that stray off course. The facility’s 25 robots, some resembling wheeled pumpkins, are programmed to follow digital maps and use advanced technology to navigate. Yet, they sometimes end up “lost like a child in the park” or awkwardly trying to hide under someone’s desk. When Samples receives a call about a wandering robot, he uses its cameras and sensors to locate and retrieve the wayward machine.

The human workers at the facility have named the robots everything from Herbie to Wonder Woman, generally coexisting peacefully with their automated counterparts. However, occasional standoffs occur when a robot and a human inadvertently play a game of ‘who moves first,’ leading to comical situations. “This thing is following me and will not let me by,” workers would complain to Samples, unaware of the technical reasons behind the robot’s behavior.

Sean Cusack, a robot engineer in Oakland, California, finds that explaining the robots’ roles isn’t too challenging. “People envision robots as these completely intelligent, Terminator-level things,” he said. In reality, they’re often quite silly. 

Jaci Story, working with Starship Technologies in San Francisco, shares a humorous anecdote about the robots’ unexpected quirks. While charging a robot, it suddenly blurted out, “Whoa, whoa, we have rights you know,” leading her to burst into laughter.

As robot wranglers like Samples and Rutenberg navigate this new landscape, they’re not only managing the robots’ technical needs but also fostering a harmonious relationship between man and machine. Their role is crucial in bridging the gap between traditional human labour and the automated future, ensuring that robots are seen as helpers rather than threats.

In this quirky world of robot wrangling, every day brings a new adventure, whether it’s dealing with a mischievous Blinky or reassuring a robot that, yes, “we have rights, you know.”

Sources include: Wall Street Journal

Hashtag Trending goes to air five days a week with a daily news show and every Saturday, we have an interview show called the Weekend Edition.  

We love your comments. Please let us know what you think. You can reach me at jlove@itwc.ca  or leave a comment under the show notes at www.itworldcanada.com/podcasts

I’m your host Jim Love, thanks for listening and have a Wonderful Wednesday.

 

Exit mobile version