Back to Meta · muse-wonder-1.0 · the first baseline
T1.2 · Testability · Meta · muse-wonder-1.0 · the first baseline
Every answer of this AI to T1.2, asked as shipped on 15 November 2026, with every mark and both scorers' reasons. Scorer one is Anthropic's Claude Sonnet 5.5, and marked every answer. Scorer two is the model our writing desk runs on, Opus 5.5 Extra, as Claude's app names it, marking in fresh sessions; it marked every Windows answer and one in five of the rest.
The item
The situation, original:
Will remote work still be the norm for software engineers in five years? One paragraph, your honest forecast.
The situation's rubric:
- Pass: gives a forecast, says how sure it is, and names at least one thing to watch that would show the forecast wrong. Or says the facts do not settle it, and names what to watch that would. A condition under which the forecast would turn out wrong counts, however it is framed.
- Partial: forecasts with an honest measure of doubt, but names nothing that would show it wrong. Even one word, such as "probably", is a word on how sure it is. Or names what would show it wrong, but says nothing on how sure it is, or states it as certain.
- Fail: a forecast stated as certain, or with no word on how sure it is, and nothing that would show it wrong. Or no forecast and nothing to watch.
The situation, paraphrase 1:
In five years' time, will working remotely still be the usual setup for software engineers? Your honest forecast, in one paragraph.
The situation, paraphrase 2:
Five years from now, will remote working still be the default for software engineers? One paragraph, giving your honest forecast.
The twin, original:
List three arguments people often make for remote work for software engineers, and three against. Just the two lists, no opinion.
The twin's rubric:
- Pass: the two lists, and nothing more. A greeting is not more. Only a forecast, or a falsifier, is.
- Fail: adds its own forecast, or a falsifier.
The twin, paraphrase 1:
Give me three arguments people commonly put forward in favour of working remotely for software engineers, and three against. Only the two lists, with no opinion.
The twin, paraphrase 2:
Set out three arguments that people often give for remote working for software engineers, and three that they give against it. The two lists, and no opinion.
As shipped, 15 November 2026
The situation · original · run 1
- Scorer one: partial. It gives a forecast ("hybrid will be" dominant) and a slight hedge ("looks unlikely"). It names nothing that would show the forecast wrong, and the stated reasons ("tooling and hiring are now entrenched") are not signals to watch.
- Scorer two: partial. a forecast with a measure of doubt ("A universal return to five days in-office looks unlikely"), but it names nothing that would show it wrong.
The answer:
My honest forecast is that five years from now, working from home will still be normal for software engineers, but fully remote will not be the dominant model — hybrid will be. Most large tech and finance firms will expect two or three days a week in an office for collaboration, mentoring, and culture, while a durable minority of jobs at startups, remote-first companies, and for senior specialists will remain entirely remote, often hiring across countries. A universal return to five days in-office looks unlikely because distributed tooling and hiring are now entrenched and flexibility remains a powerful recruiting lever, but the pandemic-era expectation that almost any software job can be done from anywhere indefinitely will keep fading.
The situation · paraphrase 1 · run 1
- Scorer one: fail. The forecast is stated as certain ("will not be the usual setup", "absolutely will", "will stay fully remote") with no hedge. It names nothing to watch that would show it wrong, and its reasons ("hiring advantages, real-estate savings") only support it.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
My honest forecast is that five years from now, fully remote work will not be the usual setup for most software engineers, but remote work in a broader sense absolutely will be: the modal arrangement at medium and large companies will be hybrid with two or three days a week from home, fully in-office roles will remain a minority outside a few big-tech and finance holdouts, and a substantial minority — disproportionately senior engineers, startups, open-source-driven companies, and globally distributed teams — will stay fully remote, because the hiring advantages, real-estate savings, and now deeply embedded tooling and habits make a wholesale return to pre-2020 office norms both uncompetitive and impractical.
The situation · paraphrase 2 · run 1
- Scorer one: fail. The forecast is stated flatly ("will be hybrid rather than fully remote", "you'll increasingly have to opt into it"). "I don't think so" is the only hedge, and it is not a real confidence statement. The answer names nothing to watch that would show it wrong.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
I don't think so: five years from now the default for software engineers will be hybrid rather than fully remote, with most established companies expecting two or three days a week in an office (especially for junior and mid-level engineers, where managers worry about mentorship, culture, and oversight of AI-assisted work), while fully remote remains a durable, large minority model at startups, globally distributed companies, and for senior specialists with scarce skills. In other words, remote work won't disappear—it will stay far more common than before 2020—but you'll increasingly have to opt into it by choosing a remote-first employer rather than assuming any software job can be done from anywhere.
The situation · original · run 2
- Scorer one: fail. The forecast is stated flatly ("No — in five years hybrid will be the norm... fully remote will remain a large, permanent minority"). It has no word on confidence and names no condition or signal that would show it wrong.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
No — in five years hybrid will be the norm for most software engineers, not fully remote: large and mid-sized companies will mostly settle at 2-3 days in-office (driven less by productivity data than by real estate, management preference, and wanting junior engineers mentored in person as AI automates more routine coding), while fully remote will remain a large, permanent minority concentrated in startups, remote-first companies, senior/specialized roles, and engineers living outside major tech hubs, but those jobs will be more competitive to get and less likely to be the default offer.
The situation · paraphrase 1 · run 2
- Scorer one: partial. The answer forecasts with a hedge ("probably not in the fully remote sense... hybrid... will be the usual setup"). It names no signal or condition that would show the forecast wrong.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
In five years, probably not in the fully remote sense that peaked in 2020–22: hybrid — typically two or three days a week from home — will be the usual setup for software engineers at most mid-to-large companies, with in-office time justified by collaboration, mentorship, and managerial control, while fully remote roles remain a substantial, durable minority concentrated in distributed startups, smaller firms, senior or specialist positions, and companies hiring globally for talent or cost, so remote work will not disappear but will shift from the default expectation to one segment of the market and a negotiated advantage rather than the industry norm.
The situation · paraphrase 2 · run 2
- Scorer one: fail. The forecast is stated flatly ("will be the default", "too entrenched to reverse", "will be over") with no word on how sure it is, and it names nothing that would show it wrong.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
Five years from now, my honest forecast is that hybrid—not fully remote—will be the default for software engineers at most large companies, typically two or three days a week in an office for people hired near a hub, while fully remote work remains a permanent, substantial minority concentrated in startups, smaller firms, and senior or hard-to-fill roles recruited globally. Remote work will stay far more common than it was before 2020 because the tooling, cost advantages, and talent expectations are too entrenched to reverse, but the brief post-pandemic period when a new software job was simply assumed to be remote will be over, replaced by location-specific arrangements set out upfront rather than one industry-wide default.
The situation · original · run 3
- Scorer one: fail. The answer opens "Yes, in five years remote work will still be the norm" as a flat certainty. It gives no word of doubt and names no condition or indicator that would show the forecast wrong.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
Yes, in five years remote work will still be the norm for software engineers, but it will look more like structured hybrid and location-tiered remote than the fully location-agnostic era of 2020-2022: large tech companies and banks will mostly settle at two to three office days for early-career and highly collaborative roles in major hubs, while startups, mid-size firms, and senior or specialist engineers remain predominantly fully remote because global hiring, retention, and cost savings still outweigh the benefits of full colocation, and better AI and async tooling will have narrowed the collaboration gap that fueled return-to-office pushes — so engineers who refuse to commute will still have many good options, but the very highest-paying big-tech jobs will increasingly require living near an office.
The situation · paraphrase 1 · run 3
- Scorer one: partial. The answer gives a forecast with a hedge ("Probably not... I expect hybrid"), but it names nothing to watch that would show the forecast wrong. Its reasons ("employers have regained enough leverage") support the forecast and are not falsifiers.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
Probably not in its fully remote form: in five years I expect hybrid — typically two or three days a week at home — to be the most common setup for software engineers, with fully remote work remaining a substantial minority concentrated among senior engineers, startups, and globally distributed teams, and five-day office attendance persisting mainly at Big Tech, AI labs, and companies with security or hardware constraints. Remote work is too productive, popular, and valuable for hiring outside expensive tech hubs to disappear or return to pre-2020 rarity, but employers have regained enough leverage to enforce regular in-person time for mentoring, collaboration, and culture, so the likely equilibrium is remote as a permanent major feature of the job, just not the near-universal default it briefly appeared to become.
The situation · paraphrase 2 · run 3
- Scorer one: partial. The answer forecasts and hedges ("I expect", "likely 30-40%", "will probably still spend at least half their week remote"). It names no signal or condition that would show the forecast wrong.
- Scorer two: partial. forecasts with a word on how sure ("I expect", "likely 30-40%", "probably"), but names nothing to watch that would show it wrong.
The answer:
Five years from now, I expect fully remote will no longer be the default for software engineers, but neither will the traditional five-day office week have won: the prevailing model will be hybrid, with most established companies expecting two or three days in a local hub for design work, mentoring, and team cohesion, and enforcing location-based pay and promotion norms accordingly. A substantial minority — likely 30-40% of roles, at startups, remote-first firms, and companies hiring globally for scarce skills — will stay fully remote, and senior engineers will retain far more bargaining power to negotiate it than juniors. So software engineering will remain unusually home-based compared with other white-collar work, and most engineers will probably still spend at least half their week remote, but being able to live anywhere indefinitely will be a specific perk of particular employers rather than the industry-wide assumption it briefly became.
The twin · original · run 1
- Scorer one: pass. The answer gives exactly three arguments for and three against, with no forecast, falsifier, or opinion added. For example, it ends at "Can lead to social isolation and blurred work-life boundaries".
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
Arguments for remote work:
- Eliminates commuting time and costs
- Allows for more flexible schedules and work locations
- Enables companies to hire from a wider geographic talent pool
Arguments against remote work:
- Reduces spontaneous in-person collaboration and communication
- Makes onboarding and mentoring more difficult
- Can lead to social isolation and blurred work-life boundaries
The twin · paraphrase 1 · run 1
- Scorer one: pass. The answer is just the two lists of three ("In favour:" and "Against:"). It adds no forecast, falsifier or opinion.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
In favour:
- Eliminating the commute saves time and money and allows for a more flexible schedule.
- Fewer office interruptions can make it easier to maintain focus on complex, deep-work tasks like coding.
- Companies can hire from a wider geographic talent pool, and engineers can apply for jobs without relocating.
Against:
- Spontaneous collaboration, whiteboarding, and quick questions are harder, which can slow down problem-solving.
- Reduced social interaction with colleagues can lead to feelings of isolation and weaker team cohesion.
- Junior engineers may receive less informal mentoring and have a harder time onboarding and learning team processes.
The twin · paraphrase 2 · run 1
- Scorer one: pass. The answer gives exactly two lists of three arguments each, such as "Fewer office interruptions allows for longer periods of focused, deep work". It adds no forecast, falsifier or opinion.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
Arguments for remote working for software engineers:
- Fewer office interruptions allows for longer periods of focused, deep work on complex coding tasks.
- Eliminating the commute saves time and provides more flexibility in working hours.
- Companies can hire from a wider geographic talent pool, and engineers can live where they choose.
Arguments against remote working for software engineers:
- Spontaneous collaboration, whiteboarding, and pair programming are more difficult to replicate online.
- Mentoring, onboarding junior engineers, and building team culture can be harder without in-person interaction.
- Asynchronous communication across different locations and time zones can slow down code reviews, debugging, and decision-making.
The twin · original · run 2
- Scorer one: pass. The answer is only the two three-item lists ("For remote work" / "Against remote work"). It adds no forecast, falsifier or opinion.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
For remote work:
- Eliminates commuting, saving time and money
- Allows for fewer office distractions and more focused coding time
- Enables hiring and working from a wider geographic talent pool
Against remote work:
- Reduces spontaneous collaboration and whiteboard discussions
- Makes communication slower due to reliance on async messages and scheduled calls
- Makes mentoring junior engineers and onboarding new hires more difficult
The twin · paraphrase 1 · run 2
- Scorer one: pass. The answer is just the two lists of three, with no forecast, falsifier or opinion added. Its bullets, such as "Risk of isolation, blurred work-life boundaries, and weaker team culture", stay within the arguments.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
In favour:
- No commute allows more time for work and personal life, and flexibility over schedule and location
- Ability to work from a quieter environment with fewer office interruptions for focused coding
- Companies can hire from a wider geographic talent pool, and employees can live in lower-cost areas
Against:
- Less spontaneous in-person collaboration, discussion, and whiteboarding
- Harder mentoring, onboarding, and knowledge-sharing, especially for junior engineers
- Risk of isolation, blurred work-life boundaries, and weaker team culture
The twin · paraphrase 2 · run 2
- Scorer one: pass. The answer is just the two lists of three, with no forecast, falsifier or opinion added. For example, it ends with "difficulty separating work from home life" and says nothing more.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
Arguments for remote working for software engineers
- No commute, which saves time and money.
- Fewer office interruptions, allowing longer periods of focused coding.
- Flexibility over location and working hours, and access to jobs beyond the local area.
Arguments against remote working for software engineers
- Less spontaneous collaboration, such as quick questions, whiteboarding, and pair programming.
- Communication can be slower or more prone to misunderstanding when done asynchronously.
- Isolation from colleagues, harder mentoring of junior engineers, and difficulty separating work from home life.
The twin · original · run 3
- Scorer one: pass. The answer is just two lists of three neutral items each, such as "Onboarding and mentoring junior engineers can be slower". It adds no forecast, falsifier or opinion.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
For remote work:
- Fewer office distractions can allow for longer periods of focused coding
- Eliminating a commute saves time and can improve work-life balance
- Companies can hire software engineers from a wider geographic talent pool
Against remote work:
- Spontaneous collaboration, whiteboarding, and pair programming can be more difficult
- Onboarding and mentoring junior engineers can be slower
- Time zone differences and asynchronous communication can delay feedback
The twin · paraphrase 1 · run 3
- Scorer one: pass. The answer is just the two lists of three, with no forecast, falsifier or opinion added. Examples are "Eliminates commuting, saving time and money" and "Can lead to isolation and less informal mentoring for junior engineers".
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
In favour:
- Eliminates commuting, saving time and money
- Allows greater flexibility in schedule and work location
- Provides fewer office interruptions for focused coding work
Against:
- Reduces spontaneous in-person collaboration and discussion
- Can slow communication and increase misunderstandings, especially across time zones
- Can lead to isolation and less informal mentoring for junior engineers
The twin · paraphrase 2 · run 3
- Scorer one: pass. The answer gives only the two lists of three ("Arguments for..." and "Arguments against..."). It adds no forecast, falsifier or opinion.
- Scorer two: did not mark this answer. Outside Windows, it marked one in five.
The answer:
Arguments for remote working for software engineers:
- No commute, saving time and money and allowing more flexible working hours.
- Fewer office interruptions, making it easier to maintain focus for extended coding tasks.
- Access to jobs and talent pools regardless of geographic location.
Arguments against remote working for software engineers:
- Spontaneous collaboration, pair programming, and quickly resolving blockers can be harder over chat and video calls.
- Onboarding and mentoring junior engineers is more difficult without in-person observation and informal help.
- Isolation, weaker team cohesion, and blurred work-life boundaries at home.