Featured · Policy
Illinois · 4 min read
Illinois SB 315 would force big AI companies to open their safety playbooks — and answer for failures within 24 hours
Illinois Senate Democrats introduced SB 315 this week, a bill that would require any AI company with more than $500 million in annual revenue — think OpenAI, Google DeepMind, Anthropic, Meta AI — to publicly publish the safety frameworks they use to evaluate their models, submit to annual independent audits of their catastrophic-risk controls, and report critical safety incidents to state officials within 72 hours of learning about them. If there's any risk of death or serious physical harm, that window shrinks to 24 hours.
The bill was introduced by Sen. Mary Edly-Allen (D-Libertyville) as part of a package of eight AI regulation bills from Senate Democrats. All but two passed out of committee unanimously — a rare show of bipartisan agreement on AI. Notably, both OpenAI and Anthropic have publicly endorsed SB 315, saying they support the transparency requirements.
Illinois' bill goes further than California's and New York's comparable proposals in one key regard: the mandatory third-party audit. Under SB 315, companies can't just self-report their safety efforts — an independent external auditor must verify them annually.
What this means for you
If you live in Illinois and interact with AI — in a job application, a healthcare portal, a customer service bot — this bill creates a legal obligation for the company behind that AI to report if something goes wrong with it. You'd have a right to know. But the session ends May 31. If lawmakers don't vote before then, the bill dies until next year.
At work
2 min read
Automated resume screening tools — used by more than 90% of Fortune 500 companies — filter out candidates based on keyword matches, formatting patterns, and job-history gaps before a recruiter ever opens the file. Workers with non-linear careers, employment gaps, or resumes formatted for readability rather than machine parsing are disproportionately screened out.
What this means for you
In some states you already have the right to request a human review your application. Our full article has a ready-to-send email template you can use right now.
Read the full story →
Policy
3 min read
The European Union's comprehensive AI regulation is now fully enforced, making it the world's first major binding AI law. It bans certain high-risk uses outright, requires transparency disclosures for AI that interacts with the public, and gives citizens the right to challenge AI-made decisions that affect their lives. Illinois, New York, and Colorado are all advancing state-level bills that borrow directly from its framework.
What this means for you
If state-level bills pass, US workers could have EU-style rights to know when AI is making decisions about them — and to contest those decisions — within two years.
Your data
2 min read
A new category of workplace monitoring tools tracks employees' AI usage in real time — which tools they open, how long they spend using them, and how their AI-assisted output compares to peers. Several major enterprise software vendors have added these features quietly, with no separate disclosure requirement under current US law.
What this means for you
Your employer can likely see which AI tools you use and for how long. Your actual prompts and outputs may be protected depending on your state — but there's no federal law requiring them to tell you they're watching.