// AI transparency

All signals tagged with this topic

Sep 19

World Model Startups Hide Development Behind Closed Doors

The world models boom is creating a new opacity problem in AI development: companies are raising massive capital on vague promises while declining to share research, benchmarks, or even basic technical details about their systems. This mirrors the pre-GPT-3 secrecy playbook but with higher stakes. If these models are truly foundational infrastructure (as investors believe), the lack of independent verification makes it impossible to assess whether the technology delivers or if it's venture-scale capital masquerading as proof.

theme-ai model evaluation capability claims AI transparency

Sep 6

OpenAI Adjusts GPT-6 Benchmarks After Launch, Raising Questions About Metric Integrity

Source: Fortune

OpenAI's post-hoc revisions to evaluation standards for its flagship model undermine the credibility of comparative AI benchmarks when governments and enterprises need reliable performance data for deployment decisions. The pattern of metric adjustment after release—rather than pre-registration—mirrors gaming dynamics in academic publishing and creates incentives for labs to optimize benchmark design around their own capabilities rather than genuine capability measurement. Benchmark gaming erodes the technical consensus required to evaluate real safety tradeoffs between models.

theme-culture regulation/policy AI transparency legal ethics

Aug 21

Lawyer Sanctioned for Hiding AI Prompts in Court Filing

Source: Boing Boing

A Connecticut judge penalized an attorney for embedding hidden prompts in court documents to manipulate how opposing counsel and the court would process the filing. The tactic exploits a gap between human and machine interpretation—adversaries can weaponize AI reading to turn document formatting into a covert attack vector. As AI becomes routine in legal workflows, courts will need explicit rules against prompt manipulation, similar to existing sanctions for obfuscation and bad faith briefing.

theme-ai model releases AI transparency security

Jul 30

OpenAI's Open-Source Security Scanner Keeps Its Core Locked

Source: The Next Web

OpenAI released Vulnerable Code Detector as open-source while withholding the AI model that performs the vulnerability scanning, making "open-source" functionally meaningless for users who can't run or audit the tool's critical component. This approach reflects a broader industry pattern: companies adopt open-source framing as marketing while keeping proprietary models that deliver value, converting transparency into branding. The gap between the licensed interface and the closed model shows how "open-source" now functions as a positioning claim in AI infrastructure rather than a technical commitment.

theme-culture regulation/policy AI transparency legal precedent

Jul 23

SpaceX's AI Lawsuit Threatens All Transparency Regulations

Source: Transformer

Elon Musk's xAI is challenging California's AI transparency requirements by arguing they violate free speech—a legal theory that, if adopted by appeals courts, could invalidate existing state laws and preempt future ones nationwide. The immediate issue is training data disclosure, but the precedent reaches further: a loss would expose transparency mandates across AI, labor, content moderation, and other industries to constitutional challenge under corporate speech protections.

theme-consumer privacy AI transparency platform trust

Jul 7

Apple warns users when AI prompts route through Google's servers

Source: AppleInsider News

Apple is applying the same transparency friction it uses for ChatGPT to Google Cloud infrastructure, turning data routing into a visible consumer choice rather than a silent backend decision. Device makers are treating AI model selection as a privacy and trust positioning lever, not just a feature, forcing both Google and OpenAI to compete on the visibility and consent choreography around where inference happens. For consumers, the question of where data goes is becoming a standard part of AI tool evaluation.

theme-culture AI transparency research integrity misinformation

May 16

EY Retracts Loyalty Study Over AI Hallucinations and Fabricated Citations

Source: Financial Times

EY's withdrawal signals that AI-generated research is entering institutional workflows without adequate guardrails, creating reputational risk even for blue-chip firms. The fake footnotes and hallucinated data points suggest researchers either didn't validate outputs or used generative AI as a shortcut to content production rather than analysis—a pattern likely replicated across consulting and professional services where speed-to-delivery pressures collide with AI's persuasive plausibility. The move will accelerate investment in detection and validation infrastructure as clients begin demanding audit trails and third-party verification of research credibility.