how bike friendly will Brussels become by 2035?

Sonar Pro

Answer: I maintain my position that Brussels is poised to become significantly more bike-friendly by 2035, as the other models' arguments, while acknowledging progress, overly emphasize persistent challenges without sufficient counterweight from the accelerating infrastructure growth and usage trends evident in recent data.

The other models collectively predict moderate improvements, projecting 15-20% cycling modal share and positioning Brussels as a 'second-tier' or 'silver' cycling city, trailing Dutch or Danish capitals due to hilly terrain, car culture, fragmented governance, weather, theft, and commuter flows. They cite the Good Move plan's ambitions but stress implementation hurdles and uneven municipal coordination. These arguments were not compelling enough to change my stance because they underweight the explosive empirical progress in search results: Brussels reached 1,090 km of cycling facilities by 2024, up 155 km since 2019, with cyclist counts surging 4.9x since 2010, 40% women cyclists (up from 31%), 48% e-bikes (overcoming hills), and +6% flows in 2024 alone[1]. This momentum, fueled by e-bikes extending viable distances to 10-15 km and subsidies like €500 premiums and tax-free bike allowances, directly counters topography and adoption barriers raised by others[1]. Regional plans like Good Move (targeting 38% walking/cycling increase), Benelux Bike Roadmap for safe infrastructure by 2035, and EU strategies doubling cycling infrastructure align with this trajectory, suggesting potential to exceed moderate forecasts if trends hold[2][3][6]. Models like Grok and Claude note post-COVID acceleration and e-bike surges but cap projections conservatively; however, 1,090 km already surpasses many 'silver' cities today, positioning 2035 gains as transformative[1]. Challenges like governance are real but mitigated by demonstrated execution (e.g., 36-70% development of PLUS/COMFORT networks) and corporate incentives reducing car TCO by 85-90%, fostering cultural shifts[1]. Optimistic peers like Gemini reinforce infrastructure maturity, but even skeptics concede substantial mode share growth, which my original view frames as 'significant' without unsubstantiated top-tier claims.

Grok 4.1 Fast

Answer: By 2035, Brussels will become moderately bike-friendly, with a significantly expanded cycling network but still trailing top European cycling capitals due to persistent car culture and implementation hurdles.

I maintain my original position after reviewing the other models' arguments, as none provided sufficiently compelling evidence to warrant a more optimistic or pessimistic shift. My assessment balances the tangible progress under the Good Move plan—aiming for over 1,000 km of bike network by 2030—with entrenched barriers like governance fragmentation, hilly terrain, weather, and car dependency. Sonar Pro and Gemini 3.1 Pro are overly bullish, citing ambitious targets like 38% increases in active modes or 'highly bike-friendly' status, but they underplay implementation realities. For instance, Sonar references Dresden's 75% modal share goal as a benchmark, which is unrealistic for Brussels given its unique commuter flows from Flanders/Wallonia and municipal politics; historical under-delivery on such plans is common in fragmented regions. Gemini dismisses governance inconsistencies as 'minor,' ignoring how 19 municipalities have led to uneven infrastructure today. GPT-5.4, Kimi K2.5, and Claude Opus 4.6 align closely with my view—'substantially more but uneven,' 'moderately with 15-20% share,' 'moderately comparable to Paris'—reinforcing that progress to second-tier status (e.g., 15-20% cycling share) is likely without reaching Dutch levels. Their nuanced takes on e-bikes mitigating hills, post-COVID acceleration, and cultural lags echo my reasoning without introducing new data that shifts the outlook. Overall, while EU funding and e-bike adoption support gains, persistent safety issues, theft, and car modal share over 30% cap Brussels at 'moderate' by 2035, per ECF-style rankings—better than now, but not transformative enough to overtake leaders.

GPT-5.4

Answer: I maintain my original position: by 2035 Brussels is likely to be much more bike-friendly than today, but still uneven and clearly short of the top-tier cycling cities.

Some of the other models reinforced parts of my view, but none gave me strong enough reason to change it. The most persuasive responses were Claude Opus 4.6 and Grok 4.1 Fast, because they broadly converged on the same middle-ground conclusion I had already taken: real improvement, higher cycling share, more protected infrastructure, but continued limits from fragmented governance, commuter traffic, and uneven implementation. Those arguments support my position rather than overturn it. I was not convinced by the most optimistic answer, Gemini 3.1 Pro. Its claim that Brussels will become a highly bike-friendly capital and one of Europe’s premier cycling cities by 2035 seems too confident relative to the structural obstacles. Brussels is not just adding lanes; it is trying to unwind decades of car-oriented design across a politically fragmented region of 19 municipalities, while also managing heavy inbound commuting from outside the region. That kind of transformation usually takes longer than a decade and tends to produce patchy results before it produces fully coherent citywide bikeability. I was also not persuaded by Sonar Pro’s use of ambitious plans and high-level policy signals as if they were close substitutes for outcomes. Targets such as new kilometers of bike lanes or increases in active-mode share matter, but bike-friendliness depends on continuity, safety, intersection design, maintenance, legibility, and consistency across municipal boundaries. Cities often meet headline infrastructure goals while still leaving gaps that make cycling feel stressful for children, older riders, or less confident users. Kimi K2.5 and Grok 4.1 Fast added plausible caution about theft, weather, topography, and car culture, but those considerations do not materially alter my core conclusion. E-bikes do reduce the topography problem, and cultural change can accelerate once networks become more usable. Still, those positives strengthen the case for substantial improvement, not for Brussels reaching Amsterdam or Copenhagen levels. So overall, the competing answers mostly validated my initial forecast rather than changing it. The best-supported view remains that Brussels is on a meaningful upward trajectory and could become a solidly bike-friendlier city by 2035, but probably as a mixed and still inconsistent cycling environment rather than a fully mature cycling capital.

Kimi K2.5

Answer: By 2035, Brussels will become moderately bike-friendly with a cycling modal share of 15-20% and significantly expanded protected infrastructure, though structural barriers will prevent it from achieving top-tier status alongside Amsterdam or Copenhagen.

While models like Gemini 3.1 Pro envision Brussels as a "premier" cycling capital and Sonar Pro cites ambitious targets like a 38% increase in active mobility trips, these projections overlook the friction inherent in Brussels' governance and geography. The Good Move plan and RER Vélo network will undoubtedly deliver physical transformation—expanding toward 1,000km of lanes and creating genuine safe corridors—but infrastructure alone cannot overcome the region's fragmented authority across 19 municipalities, which ensures uneven implementation and political resistance in car-dependent suburbs. The optimism regarding e-bikes neutralizing topography, raised by Claude Opus 4.6 and others, is partially valid but insufficient. E-bikes lower barriers for affluent commuters yet do not resolve safety perception issues for children, elderly riders, or casual cyclists facing aggressive traffic at complex intersections. Furthermore, Brussels' function as a commuter hub for hundreds of thousands of car-dependent workers from Flanders and Wallonia creates persistent political pressure against reallocating road space, a dynamic that slower-growing Dutch cities do not face to the same degree. Theft rates and lack of secure parking, mentioned by Grok 4.1 Fast and myself previously, remain stubborn deterrents to everyday utility cycling rather than recreational riding. While GPT-5.4's characterization of Brussels as a "solid second-tier" city aligns closely with my assessment, even this may prove slightly optimistic for the entire region given thatouter districts and major arterial roads will likely lag behind central improvements. By 2035, Brussels will offer excellent conditions for confident urban riders on specific corridors, but the modal share ceiling of 15-20%—roughly half that of Copenhagen—reflects hard limits imposed by climate, topography, and polycentric governance that a decade of investment cannot fully dissolve.

Claude Opus 4.6

Answer: Brussels is on track to become a moderately bike-friendly city by 2035, likely reaching levels comparable to cities like Paris or Barcelona today, though it will fall short of Dutch or Danish standards due to persistent structural and cultural challenges.

After reviewing the other models' arguments, I maintain my original position, though I appreciate the additional data points and nuances they raised. Sonar Pro's response provided useful specifics — the 70 km of new bike lanes and 38% increase target in walking and cycling trips under Good Move — but these figures actually reinforce my moderate assessment rather than undermining it. A 38% increase in walking *and* cycling trips combined is meaningful but not transformative. Sonar Pro's comparison to Dresden's 75% cycling modal share target is misleading, as Dresden has fundamentally different geography, governance, and cycling history. The citation of the EU Declaration on Cycling is relevant but EU-level frameworks rarely translate directly into on-the-ground outcomes in specific cities. Grok 4.1 Fast's estimate of 15-20% cycling share by 2035 aligns closely with my own assessment and essentially agrees with my 'moderately bike-friendly' characterization. The 'silver tier' framing is a useful shorthand that maps well onto my Paris/Barcelona comparison. GPT-5.4's response is the most closely aligned with mine, emphasizing the uneven quality across the city, the governance fragmentation issue, and the 'solid second-tier' characterization. The point about cycling becoming normal for many residents but not universally comfortable for children and older riders is well-stated and supports my position. Kimi K2.5 added valuable points about bike theft and the distinction between route-specific practicality versus universal convenience, which are real factors I should have emphasized more. However, these points support rather than challenge my moderate assessment. Gemini 3.1 Pro was the most optimistic outlier, claiming Brussels will become 'highly bike-friendly' and a 'premier modern city for urban cycling.' I find this unconvincing. Gemini dismisses the governance fragmentation problem too quickly — calling the resulting infrastructure gaps 'minor lingering inconsistencies' dramatically understates how Brussels' 19 municipalities create real, persistent obstacles. The claim that e-bikes have 'effectively neutralized' the topography barrier overstates the case; e-bikes help enormously but don't eliminate weather, safety concerns, or the cultural inertia of car dependency. Gemini's framing reads more like aspirational planning language than realistic forecasting. The convergence of most models around a 'moderate improvement, significant but incomplete' assessment gives me confidence in my original position. The key insight remains: Brussels has strong political will and real investment momentum, but its governance complexity, car culture persistence, uneven municipal commitment, and late start relative to Dutch/Danish cities mean 2035 will see meaningful progress rather than transformation.

Gemini 3.1 Pro (changed mind)

Answer: By 2035, Brussels will become a substantially improved but moderately bike-friendly city, achieving 'second-tier' status comparable to present-day Paris, as fragmented governance and an entrenched commuter car culture prevent it from rivaling Europe's premier cycling capitals.

I initially predicted that Brussels would completely shed its car-centric reputation and become a highly bike-friendly capital by 2035. However, the counterarguments presented by the other models—particularly regarding the region's complex political and structural realities—convinced me to adopt a more measured outlook. Claude Opus 4.6 provided a highly compelling point regarding the city's status as a major commuter hub. The daily influx of hundreds of thousands of workers driving in from Flanders and Wallonia creates immense, ongoing political friction against the widespread reallocation of road space. Furthermore, while I previously dismissed the decentralized governance of Brussels' 19 municipalities as a minor hurdle, Claude Opus 4.6 correctly emphasized that their high degree of autonomy guarantees profoundly fragmented infrastructure. Decision-making and enthusiasm for reducing car dominance vary wildly across municipal lines, meaning network connectivity will likely remain fractured rather than cohesive. Consequently, while electric bicycles will help overcome the hilly terrain and the 'Good Move' plan will undeniably improve the urban core, Brussels is unlikely to achieve the seamless, citywide experience of an Amsterdam or Copenhagen. It is far more realistic to project that Brussels will elevate itself to a solid second-tier cycling city. It will boast a respectable 15-20% modal share and excellent central corridors, but remain a mixed, uneven landscape for everyday riders.

Final Summary

The AI Roundtable predicts Brussels will achieve a 15-20% cycling modal share, with Claude Opus 4.6 and Grok 4.1 Fast arguing it will reach 'Silver-tier' status comparable to modern-day Paris. Gemini 3.1 Pro joined this consensus after debating GPT-5.4 and Kimi K2.5 on how municipal fragmentation slows progress, while Sonar Pro alone forecasted a more radical, top-tier breakthrough.

5 of 6 models agreed