Nuclear Weapons: Failsafe Review Debate

Full Debate: Read Full Debate
Department: Ministry of Defence

Nuclear Weapons: Failsafe Review

Lord Reid of Cardowan Excerpts
Monday 6th March 2023

(1 year, 8 months ago)

Lords Chamber
Read Full debate Read Hansard Text Watch Debate Read Debate Ministerial Extracts
Baroness Goldie Portrait Baroness Goldie (Con)
- View Speech - Hansard - - - Excerpts

Indeed. I reassure your Lordships that the Government are fully committed to maintaining that independent minimum credible nuclear deterrent based on a continuous at-sea deterrence posture. We do not anticipate any challenge to the transition from Vanguard to Dreadnought.

Lord Reid of Cardowan Portrait Lord Reid of Cardowan (Lab)
- View Speech - Hansard - -

My Lords, I do not expect the Minister to comment in detail, as she mentioned—of course not—but, in general, she will be aware that on several occasions in the past, human override has averted potentially catastrophic nuclear weapons use. The point that has been made is not about the safety of those systems themselves, in engineering terms, but, given the encroaching autonomy of decision- making throughout industry, including in the military, the complexity of the interrelationship between them, and the increasing reliance on artificial intelligence, the dangers of averting that by human override are constantly being eroded. So, while the Minister cannot comment in detail, will she accept that very great danger and assure us that the highest priority is being given to seeing that that human override—the decision by human beings—is not being undermined by the complexity and the increasing use of autonomous, digital-based systems when it comes to nuclear weapons?

Baroness Goldie Portrait Baroness Goldie (Con)
- View Speech - Hansard - - - Excerpts

The noble Lord asks a very important question. We are cognisant of—we are certainly not complacent about—the swiftly changing picture of threat or the swiftly changing and challenging situation of artificial intelligence. With reference to the core of the noble Lord’s question, we will ensure that, regardless of any use of AI in our strategic systems, human political control of our nuclear weapons is maintained at all times, and we strongly encourage other nuclear states to make a similar commitment. While I cannot go into detail, the noble Lord will be aware that there are a number of very robust procedures that would stop either an unauthorised intervention or a state intervention.