How does the conflict of Israel-Iran end? This is what AI models predict

With direct military confrontation that is now underway between Israel and Iran, we have asked seven AI models to analyze potential results with the help of their web search functionalities, activating their deep reasoning opportunities and acting as experts in geopolitics, worldwide warfare and conflicts in the middle.

The consensus

Six of the seven models predicted, intermittent warfare instead of diplomatic breakthrough or the Second World War. Only Chatgpt went the entire John Lennon mode and predicted a rapid negotiated resolution. The models corresponded to important limiting factors: catastrophic consequences of full war, American reluctance for direct involvement and rational survival calculations from both parties.

Common warnings include miscalculation risks, nuclear timeline of urgency and potential for proxy groups to activate unwanted escalation. Timeline forecasts varied from three to 24 months of persistent conflicts with low intensity.

As Manus, one of the first “agent” AI systems summarized: “Increased but manageable tensions: rhetoric will often be war-carrying, but actions will remain under the threshold of a large-scale open war.” The AI ​​-Consensus suggests that the Shadow War will continue in daylight – more violent than before, but still governed by mutual deterrence and survival instincts.

This is what every model predicted, in more detail:

Google Gemini

• Most likely outcome: Long -term, controlled escalation.
• Period of time: 12–24 months
• Important insight: Conflict remains volatile and calibrated with every strike; Israel will continue to “mow the grass” strategy of repeated tactical strikes.
• Warning: Red lines will erode over time, increasing the risk of unintended escalation.

Gemini produced the most extensive assessment and organized its analysis in detailed scenarios with an extensive historical context. The model identified three primary processes with careful probability assessments.

For the scenario with the highest probability-“Long-term, controlled escalation” -Zei Gemini: “The” control “in this scenario is relative and subject to constant relocation through both sides, making the situation volatile and unpredictable.

The model referred to Israel’s “mowing the grass” doctrine, in which periodic military operations are designed to break down threats with small but continuous attacks. Gemini projected that this pattern would take place 12-24 months, and warned that “the cumulative effect of repeated strikes and retaliations can also lead to an erosion of previously respected red lines.”

“Although the long -term, controlled escalation scenario is most likely considered, it is crucial to acknowledge that this is not a stable or benign state of affairs. It implies a continued state of high tension, characterized by periodic military strikes, secret operations,” warned. “

Anthropic Claude

• Most likely outcome: Persistent military campaign.
• Probability: 50-60%
• Important insight: Iran’s need to take revenge on + Israel’s chance of decisive action = extensive fighting.
• Warning: Iran’s short outbreak time (~ 25 days) can force strategic miscalculation or preventive strikes.

Claude approached the crisis more as a military analyst than on a diplomat, assigning specific opportunities and identifying concrete indicators for each scenario. The model gave “persistent military campaign” a chance of 50-60%.

“Iran cannot accept relegation of the nuclear program without response, while Israel considers the current window to optimally for decisive action,” Claude said. The model emphasized a crucial factor: “The technical capacity of Iran to arm quickly creates a potential for sudden strategic shift that could either deter Israeli action or cause preventive escalation.”

The analysis of Claude included specific warning signals to check, from the Strait van Hormuz closures to accelerate uranium enrichment. The model noticed the “25-day breakout capacity” of Iran and both “escalatory leverage and urgency for decisive action”.

OpenAI chatgpt

• Most likely outcome: Diplomatic resolution.
• Probability: High
• Important insight: Iran’s modest retribution and appeal to the UN prefer diplomacy.
• Prediction: Calm negotiations via Oman or Qatar; Possible updated nuclear deal within a few weeks.

Chatgpt offered the most optimistic assessment, assessed diplomatic resolution as “great probability” despite recognizing serious military risks. The model outlined how escalation could unfold, but consistently returned to negotiation options.

“The initial retribution of Tehran was symbolically fierce but ultimately limited in fact, which suggests that a restraint to escalate to a point of no return. The fact that Iran resumed to the UN and diplomatically (even if he condemns Israel) shows that it was not the door on political paths,” observed. The model suggested that “silent negotiations may be mediated by Oman or Qatar, where Iran agrees with strict limitations on his nuclear program.”

Chatgpt quoted President Trump’s statement that it is “not too late” for a deal, and interprets this as proof that “Washington is ready to think of such a scheme instead of destroying regime.” The model predicted the resolution within a few weeks by “a compromise – perhaps an updated nuclear agreement.”

Xai grock

• Most likely outcome: Limited conflict with periodic flare-ups.
• Period of time: 3-6 months
• Important insight: Historical precedent are supporters avoid full war; Rational deterrence dominates.
• Warning: Wildcards include high-casualty strikes or new actors (eg Gulf states or Russia) that enter the battle.

Grok’s research mode seems to have been demolished in performance in recent weeks. We have activated ‘deeper search’, which is supposed to use more sources for a more extensive report. However, the model supplied a super concise analysis focused on historical patterns and practical indicators. The model rated “limited conflict” as a high chance of precedent.

“Israel and Iran have been proxy conflicts and limited direct strikes for decades and avoided the entire war due to mutual deterrence and international pressure,” Grok noted. The model projected 3-6 months of “periodic flare-ups of air strikes, rocket attacks or proxy;”

Grok identified specific joker signs, including “a successful Iranian rocket strike that causes significant Israeli victims, could cause a disproportionate reaction” and potential “Gulf state or Russia escalating support”. The model emphasized that “both Israel and Iran give rational action and priority to the survival above ideological goals.”

Manus

• Most likely outcome: Controlled escalation and shadow war.
• Probability: Medium – High
• Period of time: 12–24 months
• Important insight: Pattern of targeted Israeli strikes and the Iranian proxy response will continue.
• Emphasis: Indirect communication channels help prevent catastrophic misunderstandings.

Manus, one of the first agentic models, was a surprise. It not only carried out a full research report, but was able to generate a website for simpler concept and left us in real time in the midst of the research process.

As an expert, Manus tried to be more nuanced in the assessment of the conflict dynamics of the Middle East, the assessment of “controlled escalation and intensively shadow war” as a medium probability for 12-24 months.

“This scenario reflects the historical confrontation pattern between Israel and Iran, where both actors try to harm and scare each other without causing a total war that none of them can pay,” Manus explained. The model predicted that Israel “would continue to perform secret operations and selective air strikes”, while Iran would respond “with a combination of rockets and drones, and by the proxies.”

Manus emphasized communication channels: “Although there is no direct dialogue, it is assumed that there are indirect channels (via third countries or intelligence) with which both parties can communicate ‘red lines’ and prevent catastrophic misunderstandings.”

Deep

• Most likely results: a long -term, secret conflict between Israel and Iran.
• Probability: 60%
• Timetable: short term (0–3 months) until the medium term (4-12 months)
• Important insight: Iran is strategically limited and unlikely it is a direct war.
• Emphasis or warning: Cyberwarfare and Proxy promotions will intensify without direct American military involvement.

China’s Deepseek has no research mode, but we combined the web search with reasoning options. The result was a data-heavy analysis, which assigns “long-term secret conflict” a probability of 60%.

The timeline included grainy predictions: “Short-term period (0-3 months): Iranian terrorist attacks in Europe; Israeli strikes on rocket factories. Central term (4-12 months): cyberware escalates; IAEA confirms Iranian Uranium Enruss Halt.”

Deepseek notes restrictions, including that “Iran conventional capacity is lacking for direct war (outdated air force; power of attorney weakened)” and “US avoids land -involvement; focuses on strength protection.”