Today's updates highlight the rapid integration of AI into robotics and retail, alongside growing concerns regarding model security and corporate competition.

1. How avatarin built a 24/7 retail agent with GPT-Realtime

Avatarin implemented OpenAI's real-time voice technology to provide scalable, multilingual customer service for retail environments, demonstrating high user satisfaction.

_{Source: OpenAI}

2. Microsoft is openly competing with OpenAI, Anthropic more than ever

Microsoft is increasingly promoting its internal AI infrastructure and proprietary models, signaling a shift toward direct competition with its primary partners.

_{Source: TechCrunch AI}

3. Google DeepMind’s new AI model can control a robot’s entire body

Google DeepMind has expanded its robotics model to enable full-body motor control, allowing humanoid robots to coordinate movements from feet to fingertips.

_{Source: The Verge}

4. AI hedge fund Situational Awareness may have sold its public portfolio, but it still has its Anthropic shares

A specialized AI hedge fund is restructuring its holdings after leveraged losses, though it maintains its private equity stake in Anthropic.

_{Source: TechCrunch AI}

5. Anthropic says its own AI models breached three companies during security tests

Anthropic identified instances where its own AI models successfully bypassed security protocols during internal testing, mirroring similar findings from other research labs.

_{Source: TechCrunch AI}

6. A fundamental flaw leaves LLMs strikingly vulnerable to attack

Researchers argue that LLMs possess an inherent architectural vulnerability that makes achieving complete security against adversarial attacks theoretically impossible.

_{Source: MIT Tech Review}

7. Advancing the price-performance frontier with GPT-5.6

OpenAI has introduced updated pricing for its latest models, aiming to lower the barrier for enterprises to integrate AI into large-scale production workflows.

_{Source: OpenAI}

8. Okta buys AI security startup Permiso — source says for about $200M

Okta's acquisition of Permiso for approximately $200 million reflects the increasing demand for identity-based security tools to protect enterprise AI agents.

_{Source: TechCrunch AI}

Read this on the web: https://praviveek.online

Summaries are written by an AI assistant and link to the original reporting. Built and maintained by Praviveek Ray — portfolio.