{
  "id": 16019471,
  "title": "The AI safety bottleneck: Save the world or get the bag",
  "url": "https://urgent.news/2026/10/22/the-ai-safety-bottleneck-save-the-world-or-get-the-bag",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-22T08:37:01.000Z",
  "source": {
    "name": "Business Insider",
    "slug": "business-insider",
    "url": "https://www.businessinsider.com/ai-safety-jobs-pay-work-in-labs-nonprofits-save-humanity-2026-10"
  },
  "original_language": "en",
  "account": "When Rishub Jain worked in safety at Google DeepMind, he was part of a select team dedicated to minimizing the most significant risks of AI. However, he left to establish a non-profit, Sampura Research, due to his belief that he could have a greater impact on AI safety from an external position. Jain desires the freedom to concentrate on a specific aspect of AI alignment, which he felt was overlooked. While major AI companies prioritize safety and invest in it, they could be putting more resources into it, which is what Jain found limiting. He posits that with greater investment in safety, he would still be in the lab working on his preferred research area with a larger team. DeepMind's principles emphasize a comprehensive approach to responsibility, safety, and security, but Jain's departure highlights concerns shared by other safety researchers who have left OpenAI and Anthropic. Some have warned that these companies are gambling with our lives and could develop technology that could lead to humanity's demise. The AI safety field has gained mainstream attention due to hacks by rogue agents and a series of departures from prominent companies. LinkedIn data shows a 91% increase in AI safety roles over the past year. The industry anticipates a surge in funding from forthcoming IPOs at Anthropic and potentially OpenAI. The scarcity of AI safety experts is not primarily due to finances but rather a shortage of talent. These professionals specialize in testing and training AI systems to ensure ethical operation and also focus on policy and governance. The scarcity issue is more about the limited availability of skilled personnel than financial constraints. Startups and nonprofits in AI safety face challenges in attracting talent due to the limited pool of qualified individuals. Companies like Meta and OpenAI offer substantial compensation packages to attract top experts, but some researchers are divided on whether working within tech companies to mitigate risks or stepping back to steer the technology's development toward safer outcomes is more effective. David Robinson, a former OpenAI safety team member, has stated that AI companies should implement more safety measures akin to those in airports and nuclear energy, rather than the rapid development culture prevalent in Big Tech. As the significance of AI safety grows, so has the emergence of organizations focused on training and recruiting workers in this field. These opportunities are not limited to AI experts, as people from diverse backgrounds are expressing interest in contributing to AI safety.",
  "summary": "Opportunities and challenges abound for the people working the job that could actually save humanity.",
  "key_points": [
    "Rishub Jain left Google DeepMind to focus on AI safety as a non-profit founder.",
    "Major AI companies prioritize safety but may not invest enough resources.",
    "AI safety roles increased by 91% on LinkedIn in the past year."
  ],
  "editors_take": "The exodus of safety researchers from major labs to independent nonprofits signals a growing tension between the rapid development culture of big tech and the need for specialized oversight.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}