Core employees of technology giants jointly asked: Is it time to put the brakes on the R&D pace of cutting-edge automated AI?
On July 30, core employees from nearly a dozen technology giants including OpenAI, Anthropic, and Google launched a joint action calling on the U.S. government to promote international cooperation and control the R&D pace of cutting-edge automated AI; the participants in the security debate are expanding from external critics to core employees within the industry.
Those who best understand the potential of AI are making the most prudent calls. On July 30, core employees from nearly 12 technology giants, including OpenAI, Anthropic, and Google, launched a joint action calling on the U.S. government to promote international cooperation and control the R&D pace of cutting-edge automated AI. This is not an accusation from external critics, but a self-warning from the core practitioners in the industry.
The core demands of the joint letter
The joint letter’s concerns focus on the security risks of cutting-edge automated AI: With the rapid improvement of the autonomy and automation capabilities of AI systems, its potential risk of loss of control is considered to be beyond the controllable scope of a single enterprise or a single country. Therefore, the co-signers advocate setting a more prudent pace and regulatory framework for cutting-edge AI research and development through international cooperation and government-level coordination mechanisms.
The essence of this appeal is to reset the balance between "research and development speed" and "safety and controllability". At a time when model capabilities are being iterated on a monthly basis, having people who master cutting-edge technologies take the initiative to "slow down" is of great signaling significance - it shows that the concerns within the technical circle about the risk of losing control have evolved from theoretical deductions to personal professional concerns.
From external criticism to internal awakening
From an industry perspective, this joint action continues the AI industry's long-standing debate on "safety" and "development," but the identities of the participants have undergone key shifts: in the past, it was mainly external researchers and ethics advocates who called for slowdown, but this time, the participants have expanded to include OpenAI (the company behind ChatGPT), Anthropic (Claude The company behind), Google (the company behind Gemini) and other core employees of giants. The concerns of insiders often touch industries and policymakers more than external criticism.
At the same time, the appeal targets the U.S. government and international cooperation, which also shows that safety governance is moving from corporate self-discipline to institutional arrangements at the national and multilateral levels—it is no longer enough to rely solely on the safety team within the company, and a national-level coordination mechanism is needed. Whether this joint name can be transformed into specific policies or industry self-regulatory measures is a key variable that needs to be continuously tracked in the future; under the competitive pressure of "development first", there is still no ready answer on how to find an institutional outlet for "prudentness".
Several directions worth tracking in the future:
- Policy implementation of the joint letter: Whether the U.S. Congress or regulatory agencies will promote legislative or administrative measures accordingly.
- Scope and weight of signatories: The size and level of the joint list determine the actual influence of this action.
- Internal response from the company: Whether OpenAI, Anthropic, and Google officials will respond positively to employees’ demands.
- Promotion of international cooperation: Will the AI security coordination mechanism at the multilateral level be accelerated as a result?
Reviews