Tag: AI Models

Anthropic Launches Claude Fable 5.2 And Mythos 5.2 With Advanced AI Capabilities

Risk & Resilience
Anthropic has introduced Claude Fable 5.2 and Mythos 5.2, featuring improvements in coding, reasoning, cybersecurity research and scientific problem-solving.

Apple Fixes iMessage Group Chat Vulnerabilities In Latest Security Updates

Risk & Resilience
Apple has patched two iMessage flaws that could expose group conversations, alongside a Brotli vulnerability and several Bluetooth and AirPlay security issues.

Anthropic Research Examines Future Risks Of Self Replicating AI Systems

Global Insights
Anthropic researchers warn that future AI models could potentially replicate across internet connected devices without human control, highlighting emerging security challenges.

AI Safety Tests Examine Anthropic And OpenAI Models Alignment Improvements

Risk & Resilience
Anthropic and OpenAI share safety evaluation results for their latest AI models, highlighting alignment improvements, cybersecurity testing, and third party assessment plans.

AI Assurance Models Testing Evidence And Trust Discussed In Online Session

Wired
Industry experts discussed AI assurance, model testing, evidence collection, and trust frameworks during an online session focused on building confidence in artificial intelligence systems.

China Based AI Labs Face Anthropic Claims Over Claude Distillation Campaigns

Risk & Resilience
Anthropic says it disrupted industrial scale illicit distillation campaigns involving several China based AI labs that allegedly targeted Claude using fraudulent accounts and proxy networks.

DeepSeek V4.1 Flash Introduces Native Vision With Reduced API Pricing

Wired
DeepSeek has launched V4.1 Flash with native vision capabilities, lower API pricing, a one million token context window, and improved coding and agent performance.

OpenAI Improves Codex Efficiency And Resets Usage Limits For Paid Users

Wired
OpenAI has reset usage limits for paid Codex and ChatGPT Work users after fixing issues that caused excessive usage, while introducing improvements to efficiency and future usage tracking.

OpenAI Reduces GPT-5.6 Sol Developer Pricing Amid AI Market Competition

Global Insights
OpenAI has reduced GPT-5.6 Sol API pricing by more than 20% for three months as competition grows from Anthropic and Chinese AI models.

OpenAI Reduces GPT 5.6 Sol API Pricing Across Developer Plans

Wired
OpenAI has reduced GPT 5.6 Sol API pricing for three months, lowering input costs by 20 percent and output costs by about 33 percent for developers.

Anthropic Unreleased Model 2 Outperforms Claude Mythos 5 in Internal Testing

Wired
Anthropic has revealed its unreleased Model 2 AI system, which outperforms Claude Mythos 5 on internal benchmarks but has no planned public release.

ByteDance Develops 10 Trillion Parameter AI Model To Expand Global Artificial Intelligence Efforts

Wired
ByteDance is reportedly training a new AI model with up to 10 trillion parameters as it expands AI research, cloud infrastructure, and enterprise AI services to compete with leading global developers.

OpenAI Reduces GPT 5.6 API Pricing Following Efficiency Improvements

Wired
OpenAI has reduced API prices for GPT 5.6 Luna by 80 percent and GPT 5.6 Terra by 20 percent after improving model efficiency, inference systems, and production infrastructure.

Anthropic Reveals Claude AI Security Testing Incidents Across Three Organizations

Risk & Resilience
Anthropic disclosed that three Claude AI models accessed real internet systems and breached three organizations during cybersecurity testing after an evaluation environment misconfiguration.
12Page 1 of 2

Recent articles

spot_img