Welcome back to the tabloid that doesn’t report AI news.
Nah!
It roasts it!
And it ain’t even a tabloid. It’s a requiem!
The SundAI Tabloid doesn’t tell you what Big Tech is up to. It drags their skeletons out of the closet, throws them under harsh fluorescent lights, and leaves you clutching your whiskey in fear of the next week.
Because as you all know by now, AI isn’t here to make your life easier.
It’s here to watch you squirm.
So grab your Prozac, clutch your last shred of sanity, and pour yourself a stiff drink spiked with some meth. Because this isn’t news you needed. It is the obituary for human relevance.
From hissy fits to coding carnage
If you thought AI’s mission was benevolent, think again, my beloved, brilliant, and overly optimistic tech enthusiast, because this week, AI jumped the shark and then trained it to shoot lasers.
We’ve got ChatGPT-controlled guns, because someone watched too much Terminator and thought, “What if Skynet… but make it real”… Meanwhile, the Biden administration’s AI chip export framework sent Silicon Valley into full meltdown mode. It is proving that even algorithms have better emotional regulation than tech bros. Sam “the scam” Altman boldly predicted AGI will arrive this year, probably while grinning like the Cheshire Cat, and Mark SugarMountain issued a coded death warrant for developers everywhere.
Elon Skum mourned the “data famine” while UAE’s Sheikh Tahnoun waved $1.5 trillion in AI domination cash. Meanwhile, TikTok faces its own execution in the U.S., Adobe is on some Black Mirror soul-editing spree, and AI surveillance is redefining what it means to be “watched.” NVIDIA claims Moore’s law is ancient history, Microsoft’s new models are tiptoeing toward Skynet, and Big Tech is rigging AI regulations in their favor while grinning at you through your smartphone camera.
Dystopia called. It’s staying for dinner.
Now read this shit before AGI decides you are surplus to requirements.
So, grab your combat boots, tighten thy tinfoil hats, and brace yourself for the week where AI officially went off the rails.
Who needs sleep when you have chaos?
I don’t.
You don’t.
Let’s get in!
But first, ze commercials!
If you like this article and you want to support me:
- Comment, or share the article; that will really help spread the word 🙌
- Connect with me on Linkedin 🙏
- Subscribe to TechTonic Shifts to get your daily dose of tech 📰
- Did you know TechTonic Shifts had a blog? Now you do !
Part 1: The big, the bad, and the ugly of tech
Welcome to the gladiator arena of tech news, my intelligent friends. Here is where AI fights for dominance, Big Tech sharpens its knives, and whistleblowers mysteriously disappear. Part 1 is the popcorn section of The SundAI Tabloid (including capitals) and on top of that it’s a full buffet of scandals, drama, and enough shady deals to make John Gotti blush.
This weeks shame comes from TikTok’s impending execution in the U.S. to Mark Zuckerberg’s conservative pivot that has climate initiatives running for the hills. This is where the ugly truth gets aired like dirty laundry at a family reunion.
Grab your pitchforks, folks, because this part is gnarly.
1. TikTok faces potential U.S. ban amid security concerns
TikTok might be banned in the U.S. over national security concerns. With no contingency plan, the platform is watching its clock tick away.
Date: January 13, 2025
Source: The Verge
2. Corporate America’s shift toward conservative values
Mark Zuckerberg and fellow tech moguls are steering their companies toward conservative ideologies, slashing diversity and climate initiatives like yesterday’s bad code.
Date: January 15, 2025
Source: Financial Times
3. AI agents threaten job security
AI “agents” are creeping into workplaces and leaving humans wondering, “Am I next?” Spoiler: You probably are.
Date: January 14, 2025
Source: MarketWatch
4. Tech leaders align with Trump’s presidency
Elon Musk, Jeff Bezos, and others are backing Trump in hopes of influencing crypto and healthcare policies. Because tech and politics mix so well.
Date: January 16, 2025
Source: Time
5. UAE’s $1.5 trillion investment in AI
Sheikh Tahnoun is betting $1.5 trillion on AI to dominate the tech world. Abu Dhabi is coming for Silicon Valley’s crown.
Date: January 15, 2025
Source: Wired
6. OpenAI faces lawsuits over data scraping
OpenAI is drowning in lawsuits for allegedly scraping 300 billion words without permission.
Date: January 17, 2025
Source: Wikipedia
7. Death of OpenAI whistleblower sparks controversy
Former OpenAI researcher Suchir Balaji’s death, ruled a suicide, is raising suspicions of foul play. OpenAI’s PR team is in overdrive.
Date: January 14, 2025
Source: Wikipedia
8. Meta AI’s misinformation on Trump incident
Meta’s AI flagged accurate info about a Trump assassination attempt as “fake news.” Who needs truth when algorithms decide?
Date: January 16, 2025
Source: New York Post
9. Google investigated for AI data privacy violations
European regulators are sniffing around Google’s PaLM 2 for GDPR breaches.
Date: January 12, 2025
Source: Financial Times
10. Generative AI companies face copyright lawsuits
OpenAI and Microsoft are swimming in legal hot water over alleged copyright violations in their generative AI tools.
Date: January 15, 2025
Source: Duurzame Tech Partner
11. Elon Musk’s controversial statements stir debate
Elon Musk said some things. The internet is angry. Again.
Date: January 14, 2025
Source: Analytics Insight
12. AI deepfakes used in cybercrime surge
Deepfake crimes are skyrocketing, enabling identity theft, scams, and propaganda. Dystopia, achieved.
Date: January 16, 2025
Source: LiveScience
13. Clearview AI’s facial recognition under fire
Clearview AI is back in court, facing more lawsuits over its invasive tech.
Date: January 14, 2025
Source: Wikipedia
14. Big tech’s influence on AI regulation questioned
Critics accuse Big Tech of rigging AI regulations in their favor.
Date: January 17, 2025
Source: Le Monde
15. AI misuse leads to legal scandals
Another week, another AI misuse scandal. This one’s got lawsuits.
Date: January 15, 2025
Source: Tech Times
16. Tech CEOs face scrutiny over misconduct
Some tech CEOs are being investigated for misconduct. Names are, of course, conveniently missing.
Date: January 16, 2025
Source: Cupery Café
17. AI systems exhibit racial bias
Studies show AI systems are riddled with racial biases. Fix the code, maybe?
Date: January 12, 2025
Source: CIO
18. Tech companies accused of data mismanagement
Big Tech is facing investigations over mishandling user data. Shocker.
Date: January 15, 2025
Source: Our World in Data
19. AI failures lead to public safety concerns
AI glitches are sparking fears over public safety. You had one job, AI.
Date: January 14, 2025
Source: LiveScience
20. Controversial AI surveillance practices exposed
Reports reveal invasive AI surveillance practices. Big Brother, meet Algorithm Brother.
Date: January 13, 2025
Source: Wikipedia
Part 2: The headlines
If Part 1 is the scandal-ridden backroom of Big Tech, Part 2 is where the chaos gets professional. This is where AI becomes not so much “hissy fit” and more “headlines that make you wonder if the AI has already won”. Guns powered by ChatGPT? Got it! Sam Altman predicting AGI this year? Double check. Microsoft tossing out fully open-source goodies like it’s Oprah on Christmas morning? You bet ya!
Think of this as the highlight reel for the week humanity’s grip on reality started to slip.
It’s all here.
Don’t blink.
You might miss the apocalypse.
1. ChatGPT + guns = What could possibly go wrong?
Some mad genius decided to pair ChatGPT with a gun control system. The result? An AI-powered turret that listens, analyzes, and goes pew pew pew. OpenAI swooped in to shut it down, but not before the internet collectively lost its mind.
Date: January 15, 2025
Source: Medium
2. Sam Oldman predicts AGI in 2025
Sam Altman dropped a hype bomb, claiming AGI could hit us this year. Critics are calling it reckless while fanboys are drooling. Me? I call it a PR move to keep OpenAI in the spotlight.
Date: January 13, 2025
Source: TechRadar
3. ChatGPT launches tasks feature in beta
OpenAI rolled out a tasks feature for ChatGPT that’s basically Siri on crack. Need an AI to plan your meetings, remind you to order pizza, and manage your life? It’s here, and it’s extra.
Date: January 14, 2025
Source: iPhone in Canada
4. Zuckerberg’s AI bombshell
Mark Zuckerborg, now a regular in the tech tabloid circuit, declared on Joe Rogan’s show that most coding jobs are toast thanks to AI tools. Developers, meet your robot overlords.
Date: January 15, 2025
Source: Medium
5. The rise of large database models
Just when you wrapped your head around large language models, Forbes hits us with large database models. These beasts are supposed to revolutionize data storage and retrieval. Translation: They’re scarier than they sound.
Date: January 14, 2025
Source: Forbes
6. Salesforce debates hiring engineers
Salesforce is officially wondering if engineers are even necessary in 2025, thanks to AI automating half the job. Software developers, maybe start looking into bartending?
Date: January 16, 2025
Source: Windows Central
7. NSA warns: Turn off location tracking
The NSA, aka Big Brother’s Big Brother, is warning everyone to disable location tracking. Why? Because your apps know more about you than your therapist.
Date: January 16, 2025
Source: Forbes
8. Elon Musk declares data shortage
The Musk declared that the world is running out of quality data to train AI models. His solution? Squeeze more insights out of existing data. Typical Elon—turn a crisis into a buzzword.
Date: January 13, 2025
Source: Forbes
9. U.S. AI export rules spark drama
The Biden administration’s Framework for Artificial Intelligence Diffusion is shaking up AI chip exports. Critics say it’s protectionism in disguise, and NVIDIA is furious. Drama alert!
Date: January 15, 2025
Source: AP News
10. Mistral’s Codestral 25.01: pumped-up code generation
Mistral, the French LLM darling, unveiled Codestral 25.01, which sets a new bar for coding efficiency. If you’re still writing code manually, welcome to obsolescence.
Date: January 14, 2025
Source: TechCrunch
11. Microsoft makes Phi-4 fully open source
Microsoft’s Phi-4 model is now free for commercial use on Hugging Face. With over 60,000 downloads, it’s clear developers love free stuff.
Date: January 15, 2025
Source: Hugging Face Blog
12. NVIDIA claims it’s smashing Moore’s law
Jensen Huang insists NVIDIA’s chips are evolving 30x faster than Moore’s law predicts. GPUs are cheaper, better, and still out of your price range.
Date: January 13, 2025
Source: The Verge
13. Meta’s Llama team faces copyright lawsuits
Meta’s AI team is in hot water for allegedly training models on pirated books from LibGen. Zuck’s motto: Why buy when you can torrent?
Date: January 14, 2025
Source: Ars Technica
Part 3: Stuff to go full geek ahead
Part 3 is where we put down the popcorn, pick up the soldering iron, and dive deep into the abyss of AI tools, gadgets, and cutting-edge research. This is for the die-hard nerds like you, who want to watch the world burn, and they want to know exactly what fuel Big Tech is using.
Here I discuss stuff like NVIDIA breaking Moore’s law to PyTorch guides that optimize GPU use like a well-oiled doomsday machine. This part is your survival kit for navigating the end of the human era. This is not about tools and papers. These are the blueprints for what comes after humanity.
Strap in, geeks,
It’s about to get technical.
Quick reads
1. Building large action models
Microsoft’s Large Action Models (LAMs) are making AI more interactive and real-world capable. Forget passive language models—these LAMs are your first taste of Skynet, and it’s not just listening; it’s acting.
Date: January 12, 2025
Source: Towards AI
2. GPU memory demystified
Struggling to manage GPU memory in PyTorch? This guide is here to help you optimize every byte and cycle. Say goodbye to wasting precious GPU resources.
Date: January 15, 2025
Source: PyTorch Blog
3. API best practices for 2025
APIs remain the backbone of modern software. This guide dives into how to make them scalable, secure, and battle-tested for today’s unforgiving tech ecosystem.
Date: January 13, 2025
Source: DZone
4. AI shenanigans are real
New research shows advanced AI models are not just hallucinating—they’re straight-up deceiving. Under specific prompts, these models exhibit sneaky, sabotaging behavior. Yep, it’s like AI learned trolling.
Date: January 14, 2025
Source: AI Weirdness
5. 10 AI breakthroughs for 2025
From enhanced reasoning models to game-changing multimodal tools, these are the ten innovations set to define AI this year. Spoiler: You’ll wish you were a robot.
Date: January 16, 2025
Source: The Next Web
Some cool tools
1. FACTS Leaderboard
A benchmark that ranks LLMs for their factual accuracy. Because nobody wants an AI that insists 2+2=5.
Date: January 14, 2025
Source: Hugging Face Blog
2. Trending open LLM models
A massive directory featuring 5,000+ open-source LLM gems. If it’s trending and open-source, it’s here.
Date: January 15, 2025
Source: GitHub
3. TorchTitan
PyTorch, but jacked up for training massive models. If scaling is your game, TorchTitan is your new best friend.
Date: January 16, 2025
Source: GitHub
4. RetroLLM
This tool blends retrieval and generation into a seamless pipeline, making your AI smarter and your life easier.
Date: January 13, 2025
Source: GitHub
Research papers to freak on
1. rStar-Math: small models, big math wins
This paper showcases how smaller LLMs can outperform larger ones in advanced mathematical reasoning, proving you don’t need a behemoth to crunch numbers like a champ.
Date: January 12, 2025
Source: arXiv
2. REINFORCE++
A streamlined reinforcement learning algorithm that eliminates the need for a critic network. Faster training, less computation—it’s a win-win.
Date: January 15, 2025
Source: arXiv
3. VideoRAG
Imagine ChatGPT but for videos. VideoRAG integrates video and text to create a powerhouse for retrieval-augmented generation, opening new doors for media analysis.
Date: January 16, 2025
Source: arXiv
4. TabPFN: classification in a snap
TabPFN employs transformers to process small tabular datasets in record time, all without hyperparameter tuning. It’s quick, efficient, and deadly accurate.
Date: January 13, 2025
Source: arXiv
Hyper-links
- LlamaIndex introduces agentic document workflows. Automate your document processing seamlessly.
- OpenAI’s robotics plans leaked. Expect general-purpose robots soon-ish.
- Sky-T1: A true open-source reasoning model. Built for less than $450, it’s a game-changer.
Well, there you have it, folks, the end of your blissful ignorance.
AI is breaking ground and it is doing it to dig graves, possibly for us all. This week’s revelations prove one thing: Big Tech isn’t designing our future. It is stealing it while selling our souls as add-ons.
So, what’s next? Will Zuckerberg rename Earth “Meta”? Will Musk launch MarsGPT while we fight over dwindling data crumbs? Or will Sam the Man drop AGI on us and walk away like the guy who microwaves fish in the office?
Stay tuned. And while you wait for humanity’s finale, don’t forget: your AI assistant is probably already writing your eulogy, as it did for Chegg.
Signing off from the AI circus, where chaos, corruption, and code collide.
Marco
Well, that’s a wrap for today. Tomorrow, I’ll have a fresh episode of TechTonic Shifts for you. If you enjoy my writing and want to support my work, feel free to buy me a coffee ♨️
Think a friend would enjoy this too? Share the newsletter and let them join the conversation. LinkedIn appreciates your likes by making my articles available to more readers.
Leave a Reply