Europe weighs AI dependence and self-sufficiency
Why Europe has been absent from the great AI safety debate
Saturday, 19 September 2026
Warm-up
- How do you use AI tools in your daily life?
- Should countries build their own tech or buy it? Why?
- What AI risks seem most serious to you, if any?
Vocabulary
- dilemma
- A difficult choice between options that both have problems.
- shun
- To avoid or refuse something on purpose.
- dependent
- Needing support or supply from someone or something else.
- datacentre
- A large site with many servers that store and run computing.
- self-sufficiency
- The ability to meet needs without outside help.
- resilience
- The ability to keep working or recover quickly after shocks.
- existential
- Relating to the basic survival or continued existence of people.
- supranational
- Existing above individual nations, with wider shared authority.
Reading
Europe’s AI dilemma was set out starkly this week. Christine Lagarde warned that Europeans can shun the technology and lose growth, or embrace it and become dependent on tools from the US and China. She fears being cut off unless Europe builds its own models and more datacentres, aiming for self-sufficiency.
Europe’s absence from top-level safety talks reflects its lack of AI giants like Apple, Google or OpenAI, even though the EU is powerful in regulation. Ursula von der Leyen said she will invite frontier labs to discuss pacing development, but her call drew less attention than Donald Trump’s and China’s rejection of a slowdown.
Margrethe Vestager co-authored A Transformative AI Strategy for Europe, urging investment in datacentres and resilience against “AI crises”, while pointing to the EU AI Act’s rules for “high-risk” systems. She argues that an existential threat would need a global, supranational response, and cites the UN general assembly as a venue. Adviser Frederike Kaltheuner adds that Europe already relies on US search, chips and infrastructure — dependence that will be hard to undo; and it is this dependence that limits Europe’s voice on AI safety.
Comprehension
- What two choices did Christine Lagarde set out for Europe?
- Why has Europe’s voice been weak in the AI safety debate?
- What did Ursula von der Leyen say she would do with frontier labs?
- What two solutions does Vestager’s strategy highlight?
- Which forum does Vestager cite for addressing existential risks?
Grammar focus
Adding emphasis - cleft sentences
Use a cleft to emphasise one part of a sentence: It + be + X + that/who. The cleft puts strong focus on X (person, place or idea). (Examples constructed.)
- It is **Europe that** risks dependence on US and China.
- It was **Lagarde who** warned Europeans about choosing growth or dependence.
Grammar exercise
Circle the correct option in each sentence.
- It (was Lagarde who / was Lagarde what) warned Europeans.
- It (Europe what / was Europe that) aims for self-sufficiency.
- It (was Vestager who / Vestager what) urged investment in datacentres.
- It (was dependence that / dependence what) limits Europe’s voice on safety.
- It (was Kaltheuner who / was Kaltheuner what) pointed to chips and search.
Discussion
- Should Europe prioritise self-sufficiency or global cooperation on AI?
- How much risk is acceptable when deploying powerful AI systems?
- Does strong regulation help innovation, or slow it down? Why?
- What would make Europe less dependent on US and Chinese tech?
Vocabulary, reading, and exercises with instant feedback