OpenAI Ex-Chief Scientist Ilya Sutskever Unveils  ‘Safe’ Rival AI Lab

A leading AI scientist, Ilya Sutskever, who exited Sam Altman-led OpenAI, unveiled Safe Superintelligence. The researcher is credited as the linchpin behind ChatGPT, which launched a safe AI with no ties to commercial pressures. 

The renowned researcher indicated that the newly established AI firm will prioritize safety, which he and other critics considered to be OpenAI’s blind spot. 

OpenAI Former Scientist Establishes Safe SuperIntelligence

Safe Superintelligence Inc (SSI) draws input from co-founder Daniel Gross, who previously headed AI and Apple Inc. and later at OpenAI. The two co-founders devote the company’s sole objective to advancing AI safety and capabilities, thereby avoiding the pitfalls that engulfed their immediate employer, OpenAI. 

The new firm posted a Wednesday, June 19 update on X (formerly Twitter) assuring the pursuit of safe superintelligence tapping a straight shot, goal-oriented and oriented to a single product. The SuperIntelligence promises to pursue the objective via revolutionary breakthroughs from a small cracked team.  

Sutskever’s exit from Microsoft-backed OpenAI in 2023 emerged following internal tensions regarding compromised products’ safety in pursuing profits. SSI promises not to relegate safety to the backseat, particularly by avoiding the distractions from product cycles and management overhead confronted by the larger AI companies. 

SSI emphasizes that its business model is rooted in safety, security, and process, thus insulating its operations from short-term commercial pressures. The foundation offers a sufficient shield to scale in peace. 

SSI is set to offer a purely research-oriented entity, unlike the current business model deployed by OpenAI to commercialize AI models. The firm aligns the initial product with safe superintelligence, as the name implies. 

Sutskever informed Bloomberg that by safety, they imply matching nuclear safety and not trust and safety. He added that the firm’s mission will constitute an expansion of Ilya’s devotion to “super alignment” and prioritizing human interests.

Sutskever considers the focus on purely safe development is mandatory and not the shiny products that derailed OpenAI. 

Sutskever’s SSI Prioritizes Safety and Human Interest

Sutskever explains that SSI’s AI systems would offer more general-purpose usage and expanded capabilities, unlike the present large language models. 

The primary goal of SSI is to create a safe superintelligence that cannot harm humanity. Instead, it will run guided by liberty and democratic values. 

Sutskever informed Bloomberg that the basic level of SSI’s product is safe and superintelligence ce, that it cannot harm humanity even on a large scale. 

SSI is set to have dual headquarters in the United States and Israel, and recruitment is underway. The company promises to grant an opportunity to engage in one’s life’s work and solve the critical technical challenges facing humans today. 

The development emerged after OpenAI dissolved the super alignment unit tasked with long-term AI safety. The dissolution arose following the exit of key members.  

Sutskever constituted the initial OpenAI board that sacked Altman from the chief executive role, citing safety concerns. Altman regained the post through Microsoft-aided reinstatement and appointed a board he controls. 

Altman’s return led to a new safety unit, with previous members who had been behind safety practices forced to exit or sacked. 

Leopold Aschenbrenner was a safety researcher at OpenAI. He recently criticized the company’s current security practices, describing them as egregiously insufficient. 

OpenAI Suffers Mass Exodus of Safety Team

Aschenbrenner co-founded an investment firm involved in AGI development, which was co-founded by former colleagues Gross, Nat Friedman, Patrick Collison, and John Collison. 

Sutskever exited the company alongside Jan Leike, who headed the alignment team. The latter criticized OpenAI as prioritizing profitability while compromising safety. 

Leike revealed joining Anthropic is a firm co-founded by former OpenAI researchers who exited it, citing a poor approach to safe AI. 

Sutskever and Leike’s exit led to the departure of policy researcher Gretchen Krueger, citing similar concerns. 

Former board members profiled Altman as fostering a toxic culture within the working environment. Their replacements downplayed such accusations coincidentally, with OpenAI releasing the staff from the non-disparagement agreements. 

OpenAI would succumb to public outcry to eliminate the contentious clauses from departure paperwork. 

Join the community of successful traders using Quantum Income PRO, your key to smarter investments and a more lucrative trading journey.