📰 Elon Musk Turns Twitter Into X

You know when your algebra teacher kept making you find X? Well, Elon Musk seems to be taking it way too far.

This week, the cute little bird of the Twitter logo was killed taken to live on a lovely country farm with plenty of friends, and replaced by a big old X. Even though the URL, tweet your reply button, search Twitter bar, and retweet button, have all stayed the same. Brand managers are weeping right now. 

Unfortunately, or fortunately, it's possible Musk might get into some legal trouble for the whole thing. Why? It's an X. You know, the commonly used letter, symbol, and way to sign off a text message. Also, the logo is straight-up just a Unicode character: 𝕏

Anyway, anything Musk seems to touch turns into an X. Just ask Grimes.

📰 Sam Altman’s Eyeball-scanning Crypto Project is Launched

Sam Altman, CEO of OpenAI, has now launched his eye-scanning crypto nightmare project, with eyeball scans taking place in countries including Britain, Japan and India.

Worldcoin, a novel cryptocurrency venture has secured $250 million in funding from investors like Andreessen Horowitz and Khosla Ventures.

The project employs a weirdo approach to identity verification by utilizing eyeball scanning through a device called the Orb. Users can scan their irises with the Orb, a helmet-shaped device employing infrared light to capture a unique iris pattern, generating Worldcoin cryptocurrency.

While Worldcoin holds the potential to revolutionise transactions, it has encountered criticism from privacy advocates who raise concerns about the potential tracking of individuals' movements and activities through eyeball scanning technology. Sam Altman has defended the project, assuring users that the scanning data will be encrypted, and they will have the option to opt out of tracking.

However, in Worldcoin's privacy policy, it is stated that user data may be shared with subcontractors and could potentially be accessible to governments and authorities. However, the policy also emphasizes the implementation of measures to minimize risks and protect data through encryption, thus preventing unauthorised access.

📰 Scientists Create Fake Memories Using Deepfakes

Scientists have made a groundbreaking discovery involving deepfake technology, where false memories can be implanted into people's minds.

The study, featured in the journal Nature, demonstrated how 20 participants had false memories inserted into their consciousness.

The deepfaked clips of fictitious movies proved remarkably convincing, leading participants to genuinely believe they were authentic.

The experiment even resulted in some individuals ranking the fabricated movies, which included imaginative remakes like Will Smith starring in a rebooted version of "The Matrix," as superior to the original films.

The researchers showcased other deepfake videos, from Brad Pitt in "The Shining" to Charlize Theron as "Captain Marvel,".

The deepfakes, generated using "generative adversarial networks" (GANs), a form of artificial intelligence for creating lifelike images or videos, successfully deceived an average of 49 percent of participants.

Astoundingly, 41 percent of this group even claimed the "Captain Marvel" deepfake was superior to the original.

These findings have profound implications for memory research, hinting at the possibility of creating entirely new memories or altering existing ones using deepfake technology.

This could have applications in psychological disorder treatments, like post-traumatic stress disorder (PTSD), as well as potential implications in law enforcement, where false confessions may be manufactured. However, the ethical ramifications of such technology must be critically evaluated moving forward.

📰 Publishers Like NY Times Set To Sue AI Companies For Billions

News publishers are gearing up to initiate legal action against AI firms, asserting that these companies are infringing on their copyright by utilising their content to train language models.

The dispute revolves around whether the AI firms have the right to use the publishers' content without explicit permission, while also profiting from it without offering any compensation.

In their defence, the AI firms argue that they are protected by fair use.

They contend that their usage of the publishers' content serves transformative purposes, specifically in training language models capable of generating news articles.

This has sparked a contentious legal battle between the publishers and the AI firms, leaving the future of copyright law in the age of artificial intelligence uncertain.

The outcome of these lawsuits carries significant implications for both industries. Should the publishers succeed, it could establish a precedent that restricts AI firms from using copyrighted content to train their language models, potentially hampering AI development that relies on such models.

On the other hand, if the AI firms prevail, it may pave the way for more liberal use of copyrighted material by AI entities, raising concerns about content protection for the media industry.

As negotiations continue between the parties involved, the resolution of this conflict remains uncertain.

Nonetheless, this case serves as a pivotal moment to closely examine the evolving intersection of artificial intelligence and copyright law, as it holds the potential to reshape the landscape of both the AI and media industries.

📰The New ChatGPT Update

Exciting news for ChatGPT fanboys and girls. A new feature called custom instructions has been introduced to enable you to personalise your interactions with ChatGPT, making it more responsive to your specific requirements.

This feature will be available in beta for Plus plan users and will soon expand to all users.

With custom instructions, users can now add their preferences or specific requirements that they want ChatGPT to consider while generating responses.

The best part is, there's no need to repeat instructions in every conversation. ChatGPT will remember your preferences and apply them in every subsequent interaction, making the process smoother and more efficient.

For example, a teacher creating a lesson plan no longer needs to repeat that they teach 3rd-grade science, or a developer looking for code in a language other than Python can specify it once. Even grocery shopping for a large family becomes hassle-free, as the model takes into account your need for six servings in the grocery list.

So, using this prompt:

You'll get answers like this:

Nwaah, it feels like ChatGPT is getting to know us better. We're becoming best buds.