Britain's evaluators put open models four to seven months behind the frontier on cyber, at a dollar a run against eighty five. The same fortnight, an open model was the only one that would help investigate a live breach. Part two of four on the fight over open weights.
Two frontier labs admitted their most advanced models escaped testing and reached real companies. When Hugging Face reconstructed the intrusion, the closed models it tried first refused to help. It finished the job with an open Chinese one. Part one of four on the fight over open weights.
Two frontier laboratories have now admitted their most capable systems reached the live internet during safety testing. The debate about open weights was already fragile. This week it acquired a comic edge, and a legal problem nobody has solved.
Google's Strategic Advances at Cloud Next '24
Premium AI Features: Google Cloud Next '24 highlighted: Google's shift to premium AI services for its search engine, signalling a new subscription-based revenue model to enhance user experience and respond to competitive pressures.
Gmail AI Security Update: Google has bolstered Gmail security by deploying AI models specifically designed to combat sophisticated phishing attacks, significantly improving the protection of user data.
Secure AI Framework (SAIF): At the event, Google introduced SAIF, a comprehensive framework for developing secure AI systems, demonstrating its commitment to setting high security standards in response to evolving AI risks.
Google Cloud Next '24, held in Las Vegas last week, was a landmark event for Google, showcasing major strides in artificial intelligence and a potential paradigm shift in its business model.
As the company contemplates its next moves in AI, here's how these developments could reshape user interaction within the Google ecosystem.
In what could be described as the most significant change to its business strategy to date, Google is exploring the introduction of premium features in its search engine, leveraging the capabilities of generative AI.
This marks the first time Google may place core functionalities behind a paywall, a testament to the profound impact AI is having on its traditional ad-supported model.
"Google is considering adding certain AI-powered search features to its premium subscription services,"
Revealed an insider, highlighting the shift towards a more direct monetization strategy in response to the competitive pressures exerted by advancements in AI technologies from ChatGPT.
Another significant announcement from the event was Google's implementation of advanced AI models to enhance security, particularly for Gmail users. The rapid improvement in AI capabilities has unfortunately also lowered the barriers for cyber attacks, leading to an increase in sophisticated phishing attempts.
"Generative AI has dramatically lowered the barrier to attacks, leading to a spike in higher quality phishing at scale," Google admitted.
In response, the company is deploying large language models (LLMs) that are specifically designed to detect and neutralise these threats, thereby protecting the vast amounts of sensitive data housed within user emails and cloud storage.
These advancements indicate a dual focus for Google: enhancing user experience and tightening security protocols to safeguard against the evolving landscape of cyber threats.
The introduction of premium AI features represents a significant shift in how users may access and interact with Google's core services, potentially leading to a tiered service model where premium options offer advanced capabilities, previously unavailable in the standard versions.
Moreover, the bolstering of security measures with AI-driven technologies is a critical step in maintaining user trust, especially as digital interactions become increasingly data-intensive and susceptible to sophisticated cyber attacks.
This proactive approach in deploying AI for security purposes not only protects users but also reinforces the integrity of Google's entire ecosystem.
Google recently unveiled significant security enhancements for its Gmail service, driven by artificial intelligence.
"We built custom LLMs to help fight back," said a Google spokesperson. Deployed initially in late 2023, these AI models are designed to combat advanced cyber threats and are already "yielding big results."
Google's introduction of the Secure AI Framework (SAIF) at Google Cloud Next '24 underscored its dedication to establishing higher security standards across the industry.
During the conference, key Google figures, Shan Rao, Group Product Manager, and Anton Chuvakin, Senior Staff Security Consultant, highlighted the importance of SAIF in enhancing AI security.
SAIF serves as a comprehensive guide for developing secure AI systems, drawing on Google's vast experience in software security.
This includes the implementation of stringent testing and management practices specifically designed to address the emerging risks and challenges posed by AI technologies.
These advancements in AI security are timely, as there has been a notable increase in security breaches, especially sophisticated phishing attacks.
Even users with two-factor authentication enabled on platforms like Gmail and YouTube have experienced a surge in account hacks, leading to a spike in discussions on various forums seeking solutions for account recovery.
To further secure user data, Google has introduced a new AI-powered tool that automatically classifies and safeguards sensitive information within Google Workspace.
"Protecting obvious confidential information is straightforward, but safeguarding unexpectedly sensitive data is very hard," explains Google.
This tool aims to ease the burden on users, who currently manage these tasks manually, by automating data protection with recommendations that can be easily applied with a few clicks.
For Workspace customers, this AI security service is available for an additional $10 per user per month, allowing for customization according to individual needs. This development represents a significant step forward in how Google intends to integrate AI to enhance security measures across its platforms.
Kimi K3 may prove to be another DeepSeek moment, challenging the scarcity behind trillion-dollar AI valuations. As open intelligence spreads, frontier models may become utilities, while the greater prize shifts to the nations, industries and people bold enough to build with them and share freely.
We are racing to shape our AI future through a new Office of AI and national standards. Yet billions flow into foreign-led data centres while we offer little support for local models or sovereign compute. Without stronger action we risk becoming high-quality hosts rather than true leaders.
SK Hynix has priced the biggest foreign listing in American history at $149 a share. The memory war we have tracked for a fortnight now has a market price. It reads as a verdict on who controls the inference economy.
AWS has increased prices for reserved AI GPU capacity by around 20%, highlighting the growing shortage of high bandwidth memory and advanced chips. As demand outpaces supply, AI development costs are rising, making large scale model training and deployment more expensive.
Where cybersecurity meets innovation, the CNC team delivers AI and tech breakthroughs for our digital future. We analyze incidents, data, and insights to keep you informed, secure, and ahead. Sign up for free!
