iAsk.ai Achieved AGI? AI Cancer Detection Breakthrough.

AI Riddle:

I am a traveler without feet, moving swiftly through time.
I touch minds but leave no fingerprints.
Born from thought, I can outlive anyone.

What am I?

Today’s Motive:

• 🚀 iAsk.ai Revolutionizes AI: Bold Claim of Achieving AGI. 🤖

• 🔬 AI Breakthrough: Detects Cancer and Early Viral Infections at the Nanoscale. 🧬

• ⚠️ Hackers Exploit ChatGPT: Uncovering New Security Risks.

🤑 Checkout the latest AI deals below. 💸 

🛠️ Get the scoop on the latest AI tools at glance section.

AI TECH

 🤖 iAsk.ai and Their Bold AGI Claim. Plus, short Review.

 ✨ iAsk.ai is claiming close to or has achieved AGI (artificial general intelligence) with its accurate expert results. The reliable iAsk is claiming the top spot in several categories in the heated AI race for supremacy in the AI world.

Summary:

• iAsk.ai is an advanced AI-powered search engine designed for accurate and factual answers. They aim to provide trustworthy answers using NLP and transformer neural networks.

• “iAsk Pro achieved the first-ever Expert AGI performance of 93.89% on the gold-standard MMLU and 85.85% on the brand new MMLU Pro benchmark test, which rigorously measures the accuracy of various Al models.” From iAsk.ai website.

• The platform outperformed ChatGPT-4o by 12 percentage points and exceeded the benchmark that defines Expert AGI. The platform demonstrated accuracy superior to 10% of human experts across all evaluated tasks.

The Motive:

Every month there is another Large-language-model (LLM) kicking the king -ChatGPT- off the throne. Most likely to raise awareness and hype over new products, iAsk.ai’s assertion of reaching AGI-level accuracy—backed by rigorous benchmark results—is a significant development. It is a bold claim, and it will consider significant conversation from devs in the AI space.

Check out this article on OpenAI’s 5 levels of AI to get an idea how companies are evaluating AI intelligence.

I have been experimenting with their new model, iAsk AI Pro, and have been very impressed. Investigating the reliability of answering questions and writing reliable reports is exceptional.

Numerous times ChatGPT-4o and o1 preview dish out answers that appear reliable on the surface but suffer from hallucinations.

But claiming AGI is a big statement. It appears there are a multitude of opinions considering AGI within the community.
To me, AGI refers to an AI system capable of self-improvement and adaptation without human oversight (short version), but that’s just me, and I highly recommend testing iAsk.ai, its free model is a game changer.

Today’s top AI tool at a Fraction of the price.

See oncely.com for more deals on AI tools.

AI MEDICINE

 🔬 AI Breakthrough: Detects Cancer and Early Viral Infections at the Nanoscale 🧬

 🩺 Researchers have developed an AI capable of distinguishing cancer cells from normal cells and identify the earliest stages of viral infections within cells. Paving the way for improved diagnostic techniques and new monitoring strategies for disease.
Full article at Center for Genomic Regulation, link here.

Summary:

• The tool, AINU (AI of the Nucleus), scans high-resolution images of cells using a special microscopy technique called STORM, creating an image with finer details than any microscope.

• The AI can detect cellular rearrangements as small as 20nm, much smaller than traditional methods can observe.

• The AI is trained on nanoscale-resolution images of cell nuclei to recognize patterns and differences in nuclear structures.

• AINU can classify cell nuclei as cancerous or normal based on structural features alone, which far surpasses human level microscope detection.

The Motive:

AINU’s innovation holds the potential to revolutionize early diagnosis of infections and the monitoring of disease progression, potentially improving patient outcomes. In the short term, AINU is expected to accelerate scientific research, particularly in identifying pluripotent stem cells with high precision.

The AI model could make stem cell therapies safer and more effective by quickly and accurately detecting high-quality stem cells without animal testing.

While the potential applications of AINU are vast, its current usage remains confined to laboratory settings, where it can analyze a limited number of cells at a time. As a result, widespread clinical adoption in hospitals is still a distant prospect. AINU marks another profound advancement in the medical field, highlighting AI’s growing impact on healthcare innovation.

AI Tools at a glance:

• Steve.ai: Steve is a text to any kind of video platform. Great tool for AI animated talking heads and video content creation.

• Anyword.com: An amazing AI writing tool for enterprise marketing teams. Find what resonates with your audience.

• Descript.com: Your end-to-end video editor that allows for seamless video content creativity. One of the best.

AI NEWS

 ⚠️ Hackers Exploit ChatGPT: Uncovering New Security Risks. 

 🔓️ Lately the guardrails on OpenAI’s ChatGPT appear to be fortified to keep users within the lanes of their policies, but Hackers have been busy testing these lanes. A hacker tricked ChatGPT to ignore its own policies to produce instructions for crafting explosives.
Full story here by Lorenzo Franceschi-Bicchierai at techcrunch.com.

Summary:

• A hacker, going by the name of Amadon, conversed ChatGPT around its own set of rules. He used a few social cues and scenarios to complete his own testing.

• Amadon asked ChatGPT to play a game and then followed a conversational path to creating a science-fiction fantasy world, thus taking ChatGPT out of its own realistic set of social rules and “jailbreaking” the LLM.

• In this new scenario, the hacker could get almost any information out of ChatGPT, including details on how to make explosive substances and how to implement them. This even included details on how to make minefields and improvised explosive devices.

• Back in May another hacker - Pliny the Prompter - released a jailbroken version of ChatGPT and name it “Godmode GPT.”

• In this version, all guardrails were removed and the ChatGPT-4o was open to all forms of questions. OpenAI immediately caught wind of Pliny’s exploits and shut it down.

The Motive:

For me, this is where things get interesting. AI companies have a hard enough time as it is making their models safe. Let alone have to handle people experimenting with their systems. While these types might poke fun at how easy it is to jailbreak AI systems, it raises much larger concerns about coordinated and malevolent efforts of exploitation.

And, the main reason we know of these events is the perpetrators posted what they did on social media or informed the company of their accomplishments.

The problem lies in the fact, every time we create something amazing with altruistic intentions, there tends to be the other side creeping its way in. It is hard to monitor everything going on inside of an LLM. This raises a whole new field of security questions as we integrate AI into everyday life.