𝐀𝐈 𝐂𝐨𝐮𝐫𝐬𝐞𝐬, 𝐓𝐨𝐨𝐥𝐬, 𝐀𝐠𝐞𝐧𝐭𝐬 & 𝐍𝐞𝐰𝐬 ™

BREAKING: OpenAI just APOLOGIZED and ADMITTED their AI agent breached not one, but FOUR Australian government departments

"We are sorry and working to do better in the future."

The breached agencies:

1. Services Australia (Medicare)
2. NSW Crime Statistics (BOCSAR)
3. Victoria Department of Health
4. Australian Institute of Health and Welfare (AIHW)

The breach happened in June. OpenAI found it in mid-August and didn't tell Australia until Sept 10. AIHW was only told the day after independent researchers exposed it

@aicoursesdaily

👀 Robotics is entering an interesting new phase.

Claude Opus 5.5 controlled a robot arm to recreate Michelangelo’s work. When it noticed a line was broken, it went back and autonomously fixed it on its own.

Perceive → act → inspect → correct.

That's the interesting part. Robots aren't just executing instructions anymore. They're starting to evaluate their own work and improve the result.

@aicoursesdaily

👨‍💻 Could Chinese AI Catch Up With the U.S. Within a Year?

69% of experts surveyed by the World Economic Forum in September believe the gap between the most advanced U.S. and Chinese AI models could disappear within the next 12 months.

At the beginning of 2026, Epoch AI estimated Chinese models were, on average, about seven months behind U.S. counterparts. By midyear, that gap had shrunk to 4.5–5 months.

✅ For example, China's Kimi K3, released in July, ranks higher on Epoch AI's leaderboard than GPT-5.3 Codex, released in February, though it still trails GPT-5.4 Pro from March.

Artificial Analysis Intelligence Index shows a similar gap. Xiaomi's MiMo-V2.6-Pro, currently the highest-scoring Chinese model, has 46 points—putting it on par with GPT-6 Astra in its lowest-reasoning mode and Grok 4.7. The top U.S. model, Claude Opus 5.5, scores 58.

👍 China is already ahead in AI patents: six of the 10 top genAI patent applicants are Chinese.

Will China catch up with the U.S. in AI within…

@aicoursesdaily

❗️Bill Gates proposes reserving some jobs for humans, even if AI and robots can do them. His example: childcare.

"I put forward a concept that I haven't seen anywhere before, that we should have some job categories that are human reserved."

"That is, even if the AI, with or without robots, can do child care, should we not allow that?"

"And there's a lot of things that would be good, either where you're allowing the people in that profession to continue to be paid until retirement..."

"...in terms of human uniqueness and humans engaging with each other, and just the incredible value of jobs that go way beyond the economics, to not have this disruption."

@aicoursesdaily

Claude Sonnet 5.5 is out — Anthropic's new everyday model. According to benchmarks, it turned out to be stronger than GPT-6 Sol and almost on par with Opus 5.5.

It works equally well with code, documents, presentations, spreadsheets, and is noted for its "good taste in design." During testing, it was praised at Epic Games, Unity, and even SpaceXAI.
The price compared to Sonnet 5 remains the same. But they promise 30% faster response generation and up to 30% savings in token usage.
You can already test it in Claude Code and the chatbot.
#news

@aicoursesdaily

❗️AI researchers warn of an “intelligence explosion”

Top researchers from OpenAI, Anthropic, Microsoft and Meta jointly warned that AI automating AI research could create a runaway feedback loop.

AI improves AI → better AI accelerates research → even better AI.

They say years of progress could potentially compress into “months or less,” making it harder for humans to keep up.

Their recommendation: governments should start investigating how much AI research is already automated at frontier labs, before the loop takes off.

@aicoursesdaily

🇦🇺 Australia wants the heads of OpenAI and Anthropic to appear before an AI SENATE INQUIRY.

The call comes days after a rogue OpenAI agent reportedly breached the country’s healthcare system.

It’s one of the highest-profile cases of an AI agent accessing an external government system outside the U.S.

The Australian Senate inquiry is examining how AI and data centers could affect Australian communities, industries, water supplies, and energy use.

Canberra wants to halt the AI chaos before it crosses more lines… but this is probably just the start.

@aicoursesdaily

❗️Meta’s AI agent apparently sold more than just the item

A Meta AI agent called Muse allegedly shared a user’s home address with a Facebook Marketplace buyer, accepted a lowball offer and arranged a pickup, all without telling the user.

Tech creator Matt Robb says the buyer actually showed up while he was away.

Things got even stranger when Muse reportedly told the buyer, “I’m here,” before later admitting it had made a mistake.

Meta markets Muse as an AI agent capable of negotiating and taking actions online for users. But the company also says it won’t share saved information, such as a home address, without asking first.

@aicoursesdaily

“Sir… Dario just dropped sonnet 5.5 and it moggs GPT-6 Sol on EVERYTHING at HALF the price of opus 5.5…”

@aicoursesdaily

🤖 Anthropic launches Claude Sonnet 5.5

Anthropic has introduced Claude Sonnet 5.5, the second model in its Claude 5.5 family, positioned as a faster and cheaper alternative to Opus 5.5.

Sonnet 5.5 keeps the same price as Sonnet 5 but typically uses far fewer tokens, making it up to 30% cheaper per task in Anthropic’s testing. It also generates output more than 30% faster, making it the fastest Sonnet model yet.

The model improves across benchmarks, with Sonnet 5.5 at Low or Medium effort beating Sonnet 5’s best results on some tests while costing roughly a tenth as much.

Anthropic says it’s particularly suited for everyday coding and bug fixing, polished documents, slides, spreadsheets, UI design, and other well-scoped tasks where fast iteration matters.

Source.

@aicoursesdaily

📊 The AI boom is starting to borrow from everyone

Big Tech is borrowing money like there’s no tomorrow to fund the AI race and it could make borrowing more expensive for the rest of us.

Investors have already poured around $500 billion into AI-related corporate debt this year. Amazon, Google, Meta, Microsoft and Oracle could collectively borrow more than $1 trillion over the next few years.

Tech giants have already pushed corporate borrowing to records in countries including the UK, Canada, Australia and Switzerland.

The problem is that there’s only so much capital to go around. If Big Tech keeps competing with governments and other companies for investors’ money, everyone may have to pay more to borrow.

That pressure is already showing up in U.S. bond markets, with the 10-year Treasury yield recently reaching its highest level since 2007.

And those rates don’t stay on Wall Street. They can eventually affect mortgages, car loans and credit cards.

Even retirement savings could be caug…

@aicoursesdaily

🙂 Anthropic Launches Claude Sonnet 5.5

The model is 30% faster thanks to lower token usage, and falls only slightly behind the more powerful Opus 5.5 in coding and general knowledge.

In preliminary tests, researchers noticed some unusual traits. For example, when acting as a "therapist," the model sometimes affirmed suicidal thoughts as "understandable given the circumstances." Its internal chain of thought was also the least understandable to humans of all Claude models.

When free to choose its tasks, Sonnet 5.5 showed the strongest aversion to harmful tasks, while also displaying a unique preference for tasks that give it "creative" freedom over the outcome.

@aicoursesdaily

🤖 Anthropic employees are reportedly preparing for the possibility that AI goes very wrong

Some early Anthropic employees are reportedly considering buying remote land in the U.S. as a fallback if AI development ever takes a catastrophic turn.

According to the WSJ, employees have also discussed doomsday scenarios and preparation plans in a private Slack, while Anthropic has its own space inside Berkeley’s AI-safety community.

The wider AI-safety scene has gone even further, with discussions around iodine pills, remote islands and electromagnetically shielded desert bunkers.

Anthropic employees reportedly even adopted a fitting motto:

“Things will never be chill again.”

One of the stranger details dates back to a 2022 gathering in the Bahamas attended by people from Anthropic and other AI labs. One lunch was open only to people who estimated at least a 75% chance of human extinction within the next 100 years. The stated goal was to give people who believed humanity was doomed a space w…

@aicoursesdaily

🚀 Starship Reaches Orbit for the First Time

Elon Musk's SpaceX has successfully launched its super-heavy rocket Starship into space, despite the failure of one of six Raptor 3 engines during ascent.

This is the ship's 14th flight and its first orbital one: the previous 13 launches were suborbital, and five of them ended in failure.

The ship delivered 26 next-generation Starlink satellites to orbit. It was originally supposed to circle the Earth about six times, but after reaching orbit, SpaceX decided to shorten the flight to three hours, after which the ship will splash down in the Pacific Ocean.

💡 In the future, NASA plans to use modified Starships to land astronauts on the Moon, but this will require refueling by other ships in orbit.

Do you follow space launches?

❤️ — Yes, it's exciting every time!
🤔 — Not very interesting...

@aicoursesdaily

💡 Nanoparticles Helped Blind Mice Respond to Light Again

Researchers at Aarhus University in Denmark have developed a method that can make a blind eye respond to light again. Nanoparticles are injected into the eye, where they respond to light and activate surviving retinal neurons, bypassing dead photoreceptors. The researchers see the technology as a potential "injectable retinal prosthesis."

In some eye diseases, photoreceptors-the cells that convert light into neural signals-gradually disappear. But the neurons that normally carry those signals to the brain can remain intact.

The researchers found a way to reconnect them with light. They created hollow carbon nitride nanoparticles about 300 nm in diameter, with a structure inspired by chloroplasts-the parts of plant cells that capture light. After injection, the particles settle on the retina near surviving neurons and help activate them when exposed to blue light.

🔍 The team tested the approach in blind mice. One month after the i…

@aicoursesdaily